http://www.informatik.uni-trier.de/~ley/db/conf/acl/acl2006.html
ACL06
http://acl.ldc.upenn.edu/P/P06/P06-4017.pdf	303	An Implemented Description of Japanese: The Lexeed Dictionary and the Hinoki Treebank	acquisition adding akio akira also amano analysis annotated antonyms applications approach artificial association being berlin blazing bond brant cdrom cheju chikara christine christoper coling computational concentrating conclusion conference construction contemporary cooperation corpora corpus correct coverage current currently daniel database defining delph described develop dictionaries dictionary edinburgh editor electronic english eric extended extending fellbaum first flickinger foundations francis from frontiers fujita further genres germany grammar grammars group hashimoto have hayashi high hinoki hiromi hiroshi hpsg hypernyms ijcai ijcnlp ikehara improving increasing independent information initially integrated integration intelligence international interpreted into investigating ipsj iwanami japanese joint kaname kanasugi kasahara kentaro knowledge korea korean kristina language large learning lexeed lexical lexicon linc lingo linguistically linguistics machine manning masahiro melanie meronyms miyazaki model more motivation multilingually nakaiwa nariyama natural newspaper nichols norsource norwegian oepen ogura ohtani ontology ooyama other pages paper parse parsed parsing particular precision preliminary press proceedings processing project readable redwoods references research resource robust sanae sato satoru satoshi scale scores second semantic sense sensebank sentences several shieber shigeaki shigeko shirai shoten showed siegel speech speechto springer started starting state stephan stochastic stuart such sydney tagged taikei taipei taiwan takaaki tanaka task tasks text that third this thorsten three through tokyo tomoko toutanova translation treebank treebanking trees type types understanding used useful using verbmobil verlag vocabulary volumes wahlster ways with wolfgang word wordnet work workshop yokoo yoshifumi yoshihiko
http://acl.ldc.upenn.edu/P/P06/P06-2111.pdf	257	Finding Synonyms Using Automatic Word Alignment and Measures of Distributional Similarity	acquisition aligned alon annual applications association automatic available baayen barzilay beek bilingual bouma brede brighton brown celex church clin computational computationele computer conference consortium contexts corpora corpus csail curran dagan data database della draft dyvik ecai electronic estimation europarl eurowordnet evaluation extracting extraction fellbaum finding fourth free from gertjan gosse grammatica hanks helge http ibrahim improvements information informative international itai japan jorg kathleen katz kilgarriff koehn language languages lars lecture leonoor lexical lexicography lexicon linguistic linguistics lisbon lonneke lrec machine mathematics mckeown meeting mercer ming mining mirrors moens monolingual more multilingual multilinguality mutual nederlands nederlandse netherlands networks noord norms notes nygaard optimizing opus pages parallel parameter paraphrase paraphrases paraphrasing pennsylvania people peter philadelphia philipp piepenbrock pietra plas portugal press proceedings publications references regina resource resources rijn robert sapporo schwall science second semantic semantically similar statistical stephen structural synonym synonyms syntactic taalkunde than thesaurus tiedemann toefl translation translations turney ulrike university unpublished unsupervised using versus vincent voor vossen what with word wordnet words workshop yallop zhou
http://acl.ldc.upenn.edu/P/P06/P06-2112.pdf	258	Word Alignment for Languages with Scarce Resources Using Bilingual Corpora of Other Language Pairs	aligning aswani building corpora curin data david driven english final franz gaizauskas hindi hopkins jahr john johns josef kevin knight lafferty machine melamed michael niraj noah onaizan parallel proc purdy references report robert smith statistical texts translation university using words workshop yarowsky yaser
http://acl.ldc.upenn.edu/P/P06/P06-2055.pdf	201	Analysis and Repair of Name Tagger Errors	able about accuracy achieved algorithms analysis andi anlp annotator approach arbor best bikel boundary carpuat chang chinese classification coling compared computational computationally conll cynthia daniel dekai detection difference distribution entity error errors event exceed extension extraction figure finder force formulation frequency from fung further gains gives global grishman hard heng here highperformance huang human identification improvement improving indication inference jianfeng joint judgement knowledge language learning limited linear linguistics lists lufeng major manual marine miller missing more much naacl name named names natural ning nymble only other overlap overlooked papers partially pascale pattern patterns performance picked plus pragmatic probabilistic problems proc processing programming ralph ranking rather reasoning recognition reduction reference references reflects relation resolution response richard roth rudin schwartz scott segmentation selectional sentences short shows simply single skewed some soon source sources speech spurious suggesting suggests system tagging tasks that these this through types using washington weischedel were which word workshop york zhai
http://acl.ldc.upenn.edu/P/P06/P06-1021.pdf	20	PCFGs with Syntactic and Prosodic Indicators of Speech Repairs	about acoustic acoustical acoustics additive allen america american annotated annotating annual anua applications argument arpa association atlanta bagging baltimore based basis batliner beckman benefits berkeley berlin bies bracketed breiman building buntine byrne carmichael channel chapter charniak chomsky classication cognitive communication computational conclusion conference conventional conversational core corpora corpus corrections cues database dept detection deterministic development dialog disambiguation discourse disfluencies disfluency disfluent disruptions does dorr dynamic edit edited editor editors effective electrical empirical engineering english entropy esrc event exploring exposition feature features ferguson ferrer final fluencies fong formed formedness foundations framework francisco garden godfrey gorz gregory gruyter gunther hague hale harper heck heeman help hindle hirschberg hirshberg holliman hopkins http hufnagel human hypothesis icassp icslp identifying ieee implementing improve improved incorporating input inspired institute internal international interrupted intonational isca johns johnson jorg journal juncture kahn kajarekar katz klarner labeling labelled language large learning lease levelt lickley linguarum linguistics machine macintyre march marcinkiewicz marcus markers martin maximum mcdaniel mckelvie meeting meta methods minor model modeling module monitoring mouton naacl nakatani national natural nederhof netherlands nineth noisy noord north noth ofspeech ostendorf other pages parser parsing part partially path pcfgs penn phrases pierrehumbert pitrelli predicate predictors preliminaries presented price proc proceedings processing project prosodic prosodically prosody punctuation purdue ratnaparkhi recognition recognizing references regions repair repairs report research rich roark robust royal rule rules santorini schasberger schubert science self sentence sentences sequences series shafran shattuck shriberg signal silverman simple sjlander snack snover society software sondheimer sonmez sound speaker speakers speech spilker spoken spontaneous springer standard statistical stockholm stolcke strengths structural structure structures study summer supervised switchboard syntactic syntax system tagger tagging talk tech technical technologies technology telephone their theory these thesis third tobi transcribed transcription transfer translation tree treebank tutorial understanding university using utterances uweetr variation venkataraman verbmobil verification verlag visualization volume wahlster washington weakly weintraub weischedel well weng wieling wightman wolfgang wong workshop yung zhang
http://acl.ldc.upenn.edu/P/P06/P06-2123.pdf	269	Subword-based Tagging for Confidence-dependent Chinese Word Segmentation	adaptive andi asahara azuma bakeoff barcelona chang chinese chooiling combination emerson forth fourth fukuoka haowei hongqiao huang international jeju jianfeng july kenta korea language learning machine masayuki matsumoto methods ning optimum pages proceedings processing references second segmentation sighan takashi thomas tsuzuki watanabe word workshop xinsong yotaro yuji
http://acl.ldc.upenn.edu/P/P06/P06-2105.pdf	251	A Logic-based Semantic Approach to Recognizing Textual Entailment	aaai academic again alignment allen answering applicability applications approach april august based braz canada chains challenge challenges christiane clark cogex coling collins context council countries dagan database dependency discourse dolan edinburgh editor electronic emnlp entailment europe evaluation extended fellbaum ferro form formal france from generative giampiccolo girju glickman granada guide haim harabagiu ijcai inference intelligent internatinal introduction journal kamp kluwer knowledge language lexical lexicalized logic logical made magnini maiorano manual many markert mccune member minipar model models moldovan naacl natural novischi october otter pages pair parsing pascal press proceedings processing prover publishers punyakanok question reasoning recognising recognizing reference references represent representation reyle roth salvo sammons scotland second semantic semantics some southampton spain states statistical systems szpektor table taipei taiwan tatu temporal test textual theoretic theory three time transformation vancouver ways william with wordnet workshop
http://acl.ldc.upenn.edu/P/P06/P06-2084.pdf	230	Combining Association Measures for Collocation Extraction	acquiring acquisition alberta alignment alto annual applied arbor association atoms automatic barcelona between beyond building cambridge choice choueka collocation collocational collocations comparative conf conference considerations constraints context cooccurrences corpora cost criteria cuni data databases development driven eacl edmonton emnlp empirical entire evaluation events evert exercise experiments exploiting expressions extensive extraction fawcett foundations france from graphs hastie haystack hirst http hull identification identify information inkpen interdisciplinary interesting international journal kato kita krenn laboratories language large learning lexical likelihood linguistic locating looking machine madrid manning massachusetts measures meeting methods mihalcea mining modeling models modern moore mutual naacl nagata natural near needles notes occurrences omoto order oriented pairs palmas palo parallel path pearce pecina pedersen philadelphia practical press proc proceedings processing qualitative rare ratios references regularization report representation research researchers resources retrieval retrieving riao ripley rosset saarland shimohata sigir siglex significance spain springer statistical statistics student study stuttgart sugio support suspects synonyms technical techniques testing texts textual thesis third tibshirani toulouse translation ufal univ university unsupervised using usual vector venables verlag view with word workshop yano york zhai
http://acl.ldc.upenn.edu/P/P06/P06-1120.pdf	119	Accurate Collocation Extraction Using a Multilingual Parser	academic accurate alan algorithm anthony architecture association benson better breidt brigitte budapest cambridge candidate case chapter choueka coincidence collocation collocational collocations columbus companion computational conference contentbased cooccurrences corpora cowie cruse data databases design dias dictionaries dictionary dunning editor elisabeth engineering european evert experiments expressions extracting extraction faculty feasibility firth france frequency from gael generalpurpose german handling hannah haystack honour hornby hungary hybrid identification illustrative image industrial interesting international jackendoff japan john journal justeson katz kermes krenn language large learner lexical lexicography linguistics linguistis locating looking material methods morton multiword natural needles oriented oxford pages pairs papers perspectives place press proceedings properties references rupert sapporo semantics slava some statistics stefan strevens study stuttgart surprise technical terminology text textual than thesis toulouse unit univ university user verb very volume word workshop yaacov
http://acl.ldc.upenn.edu/P/P06/P06-2054.pdf	200	Exploiting Non-local Features for Spoken Language Understanding	acknowledgements adaptive algorithms anal anonymous approach architecture assessment automatic based carnegie center chunking collective comments communication communicator comparison computational computer conditional corpora data della dialogue distant efficiently entities entropy extraction features fields finkel gibbs gildea gillick grenager helpful hidden icassp icml icslp ieee iita incorporating inducing information institute intell into issues itrc jurafsky korea labeling lafferty language large learning linguistics local mach manning marcus maximum mccallum mellon ministry model modeling models naacl page pages parsing pattern pellom pereira pietra pradhan probabilistic proceedings processing program ramshaw random recognition references relational report research reviewers roles rosenfeld sampling school science segmentation segmenting semantic sequence shallow some speech state statistical supervised support supported sutton systems technical technology text thank this three trans transformation under university using vector very ward with workshop young
http://acl.ldc.upenn.edu/P/P06/P06-2087.pdf	233	Argumentative Feedback: A Linguistically-motivated Term Expansion for Information Retrieval	aalbersberg abdou abstracting abstracts academic access acmsigir advances alexander alfonso algorithms american amia analysis anaphora annotating annu annual appear application applications approach approaches argumentation argumentative argumention aronson articles assessment assignment automatic based basis baud bibliographic biocreative bioinformatics biology biomedical biomedicine bionlp blaschke blind bowden buckley butterworths cambridge carballo case categories categorization chichester christian citations classification coling collection collections collier combination communication communications computer concept conference contextual controlled cooccurrence corpus correction critical croft database degraded demner detailed digital discovering diseases document documentation documents dumais ecir effective effectiveness efthimiadis english europe european evaluation expansion experiments extracted extracting extraction fast features feedback first flexible foundations free frei from functions furnas fushman fusion geissbuhler gene general generic genomics genre gifford gomez hall harman hersh hickam hirschman hirst hull human humphrey identification improve improving incremental informatica informatics information inquiry intensive interactive interna international into jarvinen jasis joint jones journal kaplan karlgren kaufmann keyword knowledge korhonen landauer language large latent lecture length leone lexical library linguistics lisacek loquium lynette management manning marcu marty mcknight medical medline mining mitra mizuta moens morgan mullen natural neurodegenerative nlpba normalization notes ohsumed orasan overview pages paradigm parsing patterns perez performance perret pivoted prentice press problem proc proceedings processing query rebholz references refinement related relations relevance reliable report research retrieval retrieve retrieving review rhetorical robertson rocchio ruch ruiz salton sandor savoy scale scenario schuhmann schutze science scientific search searches selection semantic sentence sentences settings sheldon shift sigir similar singhal smart smbm smith society sparck specific spelling springer srinivasan statistical stein stemming step structuring strzalkowski study suffixing summarization suppl support survey swales symp symposium system systems tanabe tapanainen tbahriti technology term test teufel text texts textual tional toward towards track transactions translation trec types understanding university using valencia variants velez veuthey vocabulary voorhees voutilainen weiss wilbur wise with word workshop zobel zone
http://acl.ldc.upenn.edu/P/P06/P06-1055.pdf	54	Learning Accurate, Compact, and Interpretable Tree Annotation	aaai accomplished accurate achieves algorithms allows also among annotations applications aspects automatic ball bank barest bayesian beating beginning behavioral best bikel bracket bracketed caraballo charniak chart chiang chomsky clustering coarse collins compact complexity computational computatoinal conclusions controls corpora data disambiguation discrimination discriminative driven each ecml enables entropy error evalb even extremely figures fine first fraction fragmentation from generalization good goodman grammar grammars grammatical hall head heads henderson hierarchical http improves inducing inference information initial inside inspired johnson klein language larger largest latent learn learned learning learns lexicalized linguistic linguistics lingusitics manning matsuzaki maxent maximum merge merging merit method metrics miyao model models more multivariate naacl natural network neural number omohundro only order outside over overcome overfitting parameters parser parsers parsing partially pcfg pcfgs pennsylvania pereira performance possible prescher press previous probabilistic program provides ranging ranks recovering reduction reestimation references refining reliably remarkably representations reranking resulting schabes schuetze science scoring sekine sense significantly sima size smaller smooth smoothing specializing split splitting statistical stolcke strategy structure summarizing symbols syntax table technique than that theory thesis this those tight training tree treebanks tsujii unlexicalized using viterbi while with without word work
http://acl.ldc.upenn.edu/P/P06/P06-1140.pdf	139	Learning to Say It Well: Reranking Realizations by Predicted Synthesis Quality	adaptive adrian alain alan algorithm alignment alyssa ambient animated annual appear arbor architecture artificial assessing assigning association atlanta australia automatically barzilay based beijing beutnagel black boosting boves bregler build bulyko caley categorial cecile ceedings characters chart china chris christian clark clickthrough clustering coling collins combinatory combining comic communication computation computational computer computers concatenative concept conference conkie context conversational coordinate corpus creating daga daniel data database davis decarlo designing dialogue directions discourse discriminative disjunctive domain doug eacl echallenges editors efficient ellen engines eurospeech exploits exploring extracting features festival fifteenth finite first forms foster freund from general generated generating generation georgia grammar graphics hands hirschberg hitzeman human hunt icassp icml icslp ijcai impact information inlg instance insuk integrated intelligence intentions international intonation intonational iordanskaja irene isca ivan iyer james janet joachims joint julia kathleen kevin king kittredge kluwer knight knowledge korin langkilde language large learning lees lenzo level lexical lidija limited linguistics logical machine mann march marcu mari marilyn mark mary matthew mckeown meaning meeting mellish michael model modeling monica multimodal multiple naacl natural next nlpxml oberlander online optimal optimizing ostendorf owen pages pang parallel paraphrase paraphrases paris parsing paul performance pitlochry pittsburgh planner planning polguere practical preferences prevost proc proceedings prosody purpose rambow raymond realization recordings regina reining representation reranking research response richard richmond robert rodriguez rogati schapire schroeter scotland scott search selection sentence sentences shimei siggraph similar simon singer speaking specifying speech spoken state statistical steedman stere stone structures stylianou swartout sydney syntax synthesis synthesiser synthesized syrdal system systems targets taylor techniques text that third thorsten towards training transactions transducers translations understand unit units using walker weighted weng white william with workshop wubin xslt your
http://acl.ldc.upenn.edu/P/P06/P06-3012.pdf	283	Focus to Emphasize Tone Structures for Prosodic Analysis in Spoken Language Generation	accent actee activity actor acts adolhosseini amsterdam analysis analyze appreciation approach aspects assume ballmer banff based beckman belgium bender boersma bought bound brennenstuhl canada cann carnegie chicago classification comm communicator complete computer concluding conf constituency constraintbased context copestake correct correspondence csli data dataset david design determine doing domain each east emily english environment evaluating ewan example extract feature festvox figure finally flickinger flower focus formal from generate grammar grammars grammatical group habilitationsschrift haji heusinger hirschberg hpsg http icslp illustrated implement implementing including information input inst interfaces intonation introduction issues ivan klein konstanz labeling language lansing lexical limited listeners machine malouf mark marks mellon methodological michigan miller minimal modify mohammad most muller multiple netherlands noaccent nobound number only orth ostendorf output pages paper parse parsed parser parses part parts patterns paul philip phonetic phonetics phonology pierrehumbert pirelli postprocessing praat press prevost price proc process produces prosodic prosody provided recursion references relate relationships remarks representation respect riehemann ronnie sciences section semantics sentence sentences series shown silverman simple since small specifying speech springer standard stanford state steedman stefan step structure study syntactic synthesis synthesized system technologies that their then theoretical theory these this thomas three tobi tone transfer translation traveling typed univ university using verbs verlag wasow weenink wightman with words york
http://acl.ldc.upenn.edu/P/P06/P06-1123.pdf	122	Empirical Lower Bounds on the Complexity of Translational Equivalence 	alignment analysis andrea andy ayan bannard barzilay based bikel bonnie burbank burch callison carpuat chiang chris christof clark clsp cohesion colin coling computational corpora corpus daniel david declan dekai description distributional divergences dorr dreyer emnlp extracting final formal from galley groups groves hall hearne heidi hierarchical hopkins html http josh kathleen keith kevin knight larger learning lexicalized linguistics link longer machine marcu marine mark markus mary mckeown melamed michel model monz naacl necip pamela parallel paraphrases parsing phrasal phrase phrases projection proposed references regina report robust rule scaling scroeder sentential shen solution statistical stephen structure transformationbased translation trees using wellington what yihai
http://acl.ldc.upenn.edu/P/P06/P06-1116.pdf	115	A Bootstrapping Approach to Unsupervised Detection of Cue Phrase Variants	aaai aaron abstracts academic accurate acts agichtein agnes alignment annotation answering application approach argumentative articles automated automatic automatically barzilay based beckwith biomedical bootstrapping briscoe british burnard butterworth carroll chen chichester chin choice chris christian christiane christine classification clustering cognitive collections collocation conference constructing context cornelis corpora corpus deepak derek dictionary digital discourse discovering diseases distributional divergence document donald eaclworkshop edinburgh edition editors eduard edvard ellen emnlp english entropy essen eugene evaluation evelyn expansion extracting extraction facts fellbaum fernando five francine frederique from general generation george gravano greg gross guide guido hindle hovy hyland icassp identification ieee indexing indicating information international interpreted jacquemin jianhua john joost journal judith julian kaplan katherine klavans kupiec laboratory language lapata large learning lexical libraries lillian linc linguistically lisacek literary london lrec luis markers measures meta metadiscourse miller ming minnen mirella modelling models morphology multi multiple myers naftali national neurodegenerative norman noun occurrence oddy oxford pado paice paper papers paradigm parsed parser patterns pedersen pereira persuasion phrases plain pragmatics predicateargument princeton problemstructuring proc proceedings program question ravichandran reference references regina relations report research retrieval richard rijsbergen riloff robert robertson robust sandor school science scientific sebastian self semantic sequence shannon shift sigir similarity simone smbm smoothing snowball space speech statistical steinbiss stephen stochastic structure structures summarist summarization summarizer surface synonymous syntax system tasks technical terms teufel text theory thesis this tipster tishby trainable transactions translation tzoukermann university users using version volker williams word wordnet words workshop zhou zoning
http://acl.ldc.upenn.edu/P/P06/P06-2008.pdf	154	Towards Conversational QA: Automatic Identification of Problematic Situations and User Intent 	amit analysis andrew annual answering arbor association automatically bill boni burger cardie chaudhri chin christian claire clarification computational corrections david dialogue dialogues diane document ellen experiments gaizauskas george harabagiu hickl hirschberg identifying interactive israel issues jacquemin john julia june lehmann linguistics litman maiorano manandhar marc marco meeting michigan miller moldovan naacl nist ogden pages prager proceedings program question ralph references research riloff roadmap robert rohini sanda shrihari singhal spoken steve structures strzalkowski suresh swerts systems tasks tomek user vinay voorhees weishedel with
http://acl.ldc.upenn.edu/P/P06/P06-2118.pdf	264	Aligning Features with Sense Distinction Dimensions	annual antal arbor artificial assoication avrim barcelona blum bosch cambridge chen chia chinese ching chiou christiane city clustering coling collocation computational conference daelemans dang database disambiguating disambiguation discrimination dong eduard electronic english entropy examples features fellbaum fourth fudong gaussian hendrickx high hoste hovy hownet http integrating intelligence international into investigations iris island issue jeju jinying joint july keenage korea lance langley language learning lexical linguistic linguistics machine marcus martha maximum meeting michigan modals model mtchchell naacl natural nineteenth ontonotes optimization palmer parameter pennsylvania performance press prior proc proceedings processing qiang ralph ramshaw references relevant report rich robust role roles ronald rosenfeld selection semantic semantics sense senses sept sighan simple smoothing solution spain special stanley systems taipei technical thesis towards university using verb verbs veronique walter wanyin weischedel wenjie with word wordnet workshop york zhendong
http://acl.ldc.upenn.edu/P/P06/P06-1132.pdf	131	Learning to Predict Case Markers in Japanese	alignment analysis anlp annotated assigning automatic baldwin bank based blaheta borrowed bunpou cambridge carreras case charniak cherry clarkson clause coling collins computational conll construction constructions context corpus corston data dependency dictionary discriminative esca eurospeech frame from function fundamental gamon generation german gildea grammar hacioglu haghighi icml ieee improved improves informed interpretation introduction isahara japanese joint jurafsky kaji kawahara keywords kingsbury kiso kurohashi kuroshio kyoto labeling language learned learning linguistic linguistics machine machines making manning martin masuoka meaning menezes method modeling models moore murata naacl nagao natural nihongo oliver operations palmer parsed parsing phrasal pradhan proceeding proceedings project proposition quirk realization references relative reranking revised ringger role roles rosenfeld rquez sekine semantic sense sentence shallow shared shintakusu shuppan statistical structure supervised support syntactically syntax tags takubo task teramura text that tokyo toolkit toutanova translation tree trees uchimoto university unsupervised uses using vector version volume ward workshop
http://acl.ldc.upenn.edu/P/P06/P06-2103.pdf	249	Discourse Generation Using Utility-Trained Coherence Models	academic adjoining alexander althaus applications approach aravind artificial barbara barzilay based best better brown building carlson catching centering chiang coherence coherent computational computing content corpus current david della dialogue directions discourse drift elhadad entity ernst estimation forbes framework generation grammar grosz hltnaacl huang inferring information intelligence international iwpt joshi journal karamanis kathleen kluwer koller kuppevelt lapata lexicalized liang lillian linguistics local locally ltag machine marcu mathematics mckeown mercer miltsakaki mirella modeling models multidocument news nikiforos noemie okurowski ordering pages parameter parsing peter pietra prasad probabilistic proceedings publishers references regina research rhetorical robert sarkar scott semantics sentence smith statistical stephen strategies structure summarization system tagged technologies theory translation tree vincent webber weinstein with workshop
http://acl.ldc.upenn.edu/P/P06/P06-1098.pdf	97	Left-to-Right Target Generation for Hierarchical Phrase-based Translation	alfred arbor assembler based brown chiang comput computational david della directed estimation hierarchical jeffrey june linguistics machine mathematics mercer michigan model pages parameter peter phrase pietra proc pushdown references robert statistical stephen syntax syst translation translations ullman vincent
http://acl.ldc.upenn.edu/P/P06/P06-1102.pdf	101	Names and Similarities on the Web: Fact Extraction in the Fast Lane	aaai academic agichtein among anlp annual answering antonio applied approach arbor artificial asked association athens automatic barcelona based before bootstrapping boston brants building cafarella canada carbonell classification clustering coling collection collections college collins columbus computational conference contexts corpora data development dictionaries digital discovering discovery distributional downey driven echihabi emnlp empirical english entities entity etzioni explorations extracting extraction fast fleischman florida from gravano greece greenwood grefenstette grishman hasegawa hindle hovy human induction information instance intelligence international japan jones kluwer knowitnow language large learning level lexicons libraries linguistics lita maryland massachusetts meeting method methods michigan models montreal multi named national natural noun offline ohio online orlando pages park part pattern pennsylvania pereira philadelphia pittsburgh plaintext predicateargument proceedings processing publishers quebec question questionanswering questions references relations research retrieval riloff sapporo scalable seattle sekine semantic sigir similar singer snowball soderland spain speech statistical stevenson strategies structures tagger technology test texas thelen thesaurus they tice tishby unsupervised using vancouver very voorhees washington words
http://acl.ldc.upenn.edu/P/P06/P06-2019.pdf	165	Constraint-based Sentence Compression An Integer Programming Approach	access accurate akira algorithms allows also ambiguity amount anlp annie annotation another appealing apply approach arbor artificial aspect automated automatic barcelona based bases best beyond boston brian briscoe brooks cambridge canada canaria carroll charniak chiori clarkson clauses cole coling college comparable compression compressions computing concerns conclusions condensation conll considering constraints corpora could crouch daniel data decoding demonstrate determining direction disambiguation discourse discrete discriminative discussed document does dolan domains each eacl edmonton eliminating employed english eugene eurospeech evaluation extraction fashion flannery flexible formulation framed from fukushi function functional furui future gains general geneva global globally grammar gran greece have here highlight holds hongyan hori horiguchi hybrid ieice important includes incorporation index inference inferring information instead integer intelligence into introduction isolation issue italy jenine jing kevin king knight knowledge labeling language languages large latter lean learning less level lexical linear linguistically lrec machines mainly marciniak marcu masaru mathematical mcdonald method methods michael minh model modeling models more motivated munirpallam naacl natural nguyen novel numerical objective observe obtain olivers optimal optimization order other packing pages palmas paper parallel park performance philip pipeline plan portability possible presence presented press probabilistic proceedings process program programming promise punyakanok recipes reduction references relatively relies require results rhodes richard riezler robust role ronald rosenfeld roth ryan sadaoki saul scale scientific score search searching seattle semantic sentence sentences shimazu shown significance significant similar simple small soft space spain speech state statistical stefan stochastic strube subordinate subtitling such summarization supervised supervision support susumu switzerland syntactic systems tasks terms teukolsky text than that this through tomasz toolkit topical tracy training transactions trento turner university unsupervised using vandeghinste vasin vector venkataramanan vetterling vincent wayne when whether whose william winston with without word words work workshop would yields zaenen zimak
http://acl.ldc.upenn.edu/P/P06/P06-1025.pdf	24	Dependencies between Student State and Speech Recognition Problems in Spoken Tutoring Dialogues	aaai abella acoustic annotating annoyance architecture artificial assignment atlas audiovisual automatic barry based between bigrams blame boland bratt bulyko certainness chase chat clark coach cobotds combining communication communicator computer conference conversational corpus correction corrections cues darpa detecting detection developing dhillon dialog dialogue dialogues discourse effectiveness emotional emotions engineering error errors essay eurospeech evaluating evaluation exploring failures features feeling forbes frampton frustration gabsdil general generation goldberg hirschberg howe human hyperarticulated icassp icslp identify ijcai implications intellig intelligence intelligent interactions interspeech isbell jordan journal kamm kearns kirchhoff know knowing krahmer krupski language large last learning lemon liscombe litman made memory models national natural ostendorf other paradise passonneau performance peters physics practical pragmatic predict problems proc prosodic prosody qualitative quantitative reasoning recognition recognizers recovery references reinforcement relationships response responses riley rotaru schultz scot shriberg sigdial singh skantze soltau specialized speech spoken states stolcke strategies student swerts system systems towards tutor tutorial tutoring usability user using vanlehn venditti vocabulary waibel walker with workshop writing
http://acl.ldc.upenn.edu/P/P06/P06-1115.pdf	114	Using String-Kernels for Learning Semantic Parsers	aaai advances alex alexander algorithm also american amounts andreas annual another appear approach arbor armanasu artificial association atis august austin available bartlett based being bernhard best between binary bobrow both build calculates cambridge capable categorial challenges character chen chris city classification classifiers clause closeness codes coling collins communications compares comparisons complete composing compositionally computational computes computing conclusions conf conll constructors context conversational corpora corpus correcting corresponds corrupted craig cristianini cruz dale darpa database databases david degrade degrades difference different distance doklady domain each earley edinburgh edit editors efficient etzioni european evaluated evaluation experiment experimental faster favorably february figure form formal free freiburg frequency from fully geneva germany glass gracefully grammars harmonic http huma human hypothesis ieee increasing inductive insertions integrates intelligence interfaces intl introduction james john journal july june kate kernel kernels language languages large later learn learning level levels levenshtein likelihood linguistics lodhi logic logical luke machine machinery machines make manual margin maria mean meaning measure meeting methods michael miller modern mooney much multiple naacl natl natural nello noise north november only oren original other outputs padd pages parameter parameters parse parser parsers parsing patti pdrop performance phonetics physics pittsburgh platt popescu portland precision prefix presence presented press price probabilistic probabilities probability proc productions programming proportional qualitatively queries rather raymond real recall references regularized report representation representations research results reversals richard robert robocup rohit santa saunders scholkopf schuurmans schwartz scotland scott seen semantic semantics sentences server shawe should showed showing shows significant similar smola soccer sourceforge soviet speech spoken stallard statistical stolcke string structured subsequence substituting support switzerland syntax system systems tang taylor technical technology test texas text than that then these this thus tractability trained training transform translation type ulate uncertainty uniformly university used users using varied vector verifying version victor volume watkins were when where which while with wong word words workshop world yates york zelle zettlemoyer
http://acl.ldc.upenn.edu/P/P06/P06-3010.pdf	281	A Hybrid Relational Approach for WSD ­ First Results	academic agirre algorithms amta appear applications approach berlin britain brown bruce classification coling combining cross data disambiguation dorr edmonds genus great guthrie http hutchins impact industrial introduction katsova kluwer knowledge kros langhorne language lexical machine management mining missing nantes performance press proceedings references selection sense senseval somers sources stevenson study systems translation trends weighted with word wordnet
http://acl.ldc.upenn.edu/P/P06/P06-1114.pdf	113	Methods for Using Textual Entailment in Open-Domain Question Answering	abdessamad alignment answering approach association barzilay bernardo burger challenge challenges channel computational corpus dagan daniel echihabi empirical entailment equivalence ferro from generating glickman headlines john learning lexical lillian linguistics lisa magnini marcu meeting model modeling multiple naacl news noisy occurrence oren pages paraphrase pascal probabilistic proceedings question recognizing references regina semantic sequence setting textual unsupervised using work workshop
http://acl.ldc.upenn.edu/P/P06/P06-1054.pdf	53	A Fast, Accurate Deterministic Parser for Chinese	algorithm analyzer andy antal applied asahara asian augmented based benfeng bikel boosting bosch brill chen cheng chiang chinese coling collins combining daelemans daniel data david dependency deterministic diversity driven emnlp entropy eric experiments exploiting freund fung generative grace guide head hearne henderson iccc icml ijcnlp information jakub john language latent learning lexicalized machine mary masayuki matsumoto maximum michael models natural ngai oriented parameter parser parsers parsing pascale penn pennsylvania proceedings processing recovering reference references report robert schapire second sloot space statistical structure technical thesis tilburg timbl transactions transformation treebank treebanks university version walter with workshop yang yoav yongsheng yuchang yuji zavrel
http://acl.ldc.upenn.edu/P/P06/P06-1036.pdf	35	Enhancing electronic dictionaries with an index based on associations	adam agirre alberto algorithm annual ansa applications applied artificial association associations avancini based basis beeferman behaviour berger bernardo bilac brno brown categorization church coling computational computing concepts conference customizations david definitions description dictionaries dictionary domain dominique doug durgar dutoit ecai eduard eneko enhancing enriching european expanding extensions fabrizio find from geneva global hanks hashimoto henri hovy hozumi http idea ilknur information intelligence journal kahlout kemal kenneth lafferty language lavelli learning lexical lexicography lexicons linguistics lookup lyon machine magnini martinez mcneill meanings meeting michael models mutual naacl natural network norms nugues oflazer olatz other pages patrick phenomenon pierre press processing references resources retrieving roadmap roberto roger search sebastiani segmentation signatures slaven specific statistical symposium taiichi takenobu tanaka target tenth term text their tokunaga tongue topic upenn using verbal watanabe wataru with word wordnet words workshop zanoli zock
http://acl.ldc.upenn.edu/P/P06/P06-2036.pdf	182	Factoring Synchronous Grammars By Sorting	above albert algorithm algorithms atkinson based bilingual binarization cambridge charles chiang cliffs compiling complexity composi computational conclusion context cormen corpora daniel david dekai emnlp englewood enoch enumeration factored free galley gildea giorgio grammars hall hierarchical hopkins huang identifies integer into introduction inversion jeffery journal kenji kevin klazar knight leiserson liang linguistics machine marcu mark melamed michel model multitext naacl only pages parallel parsers parsing permutation permutations peserico phrase prentice press proceedings references results rivest ronald rule satta sequences simple some statistical stochastic synchronous syntaxbased theory thomas transduction translation ullman volume what whether yamada zhang
http://acl.ldc.upenn.edu/P/P06/P06-1064.pdf	63	Creating a CCGbank and a wide-coverage CCG lexicon for German	abeille acquiring acquisition adjoining alena amit andy anne annotated annotation annual aoife arbor association automated automatic baldridge bangalore banks barbora barcelona based blackwell bohomva brants building cahill cakici carpenter categorial chapter chen christian clark cleaner combinatory compact computation computational conference control corpora curran dependencies dependency derivational dipper donovan dubey edinburgh editor editors engineering english erhard evaluation ewan extraction formal forst frank from gazdar genabith generalised generative geoffrey george gerald german gram grammar grammars hajic hajicova hansen head hinrichs hladka hockenmaier implementation induction informatics international ivan james japan jason john josef journal julia june keller klein kluwer language level levine lexical lexicalized lexically lexius linear linguistic linguistics lrec mairead mark martin mccarthy meeting models multilingual natural oxford pages palmas parsing philadelphia phrase prague predicative press probabilistic proceedings pullum references research resources robert rohrer roth ruken rules ruth sabine sapporo scenario school shanker silvia sister smith sozpol spain specified srinivas statistical steedman stefanie stephen structure student syntactially theories theory thesis third three tiger tree treebank treebanks turkish unification university using vijay with wolfgang workshop
http://acl.ldc.upenn.edu/P/P06/P06-1063.pdf	62	QuestionBank: Creating a Corpus of Parse-Annotated Questions	absolute accuracy achieves acquired adding against algorithm andy annotated antecedents aoife approximations atis automatically barcelona based beatrice bergen beta bikel bound building burke cahill change charles clark classifiers coling collins computational conclusions corpus curran currently daniel darpa data dekai dekang dependency design diego distance doddington does domain donna donovan driven editors effort emnlp empty encouragingly engine english evaluate evaluating extraction figure from gain genabith george gildea godfrey grammars harman have head hemphill hidden high implies induced james john johnson josef judge july justify language large learned learning lillian lingual linguistics little long marcinkiewicz marcus mark mary matching method michael mitchell models more multi natural nodes norway object output pages parallelprocessing parser parsing part pattern pcfg penn pennsylvania performance philadelphia pilot pittsburgh precision proceedings question questionbank quite reached recall recovering recovery references release represented resolution resolved resources roth ruth same santorini score scores sentences show shows simple small spain speech spoken statistical steedman stephen strong sufficiently system systems table taipei taiwan that their thesis this training treebank trees university upper using valley variation very while widecoverage will with working workshop
http://acl.ldc.upenn.edu/P/P06/P06-1108.pdf	107	Event Extraction in a Plot Advice Agent	abney american analysis andrei automatic autotutor baldwin bartlett bear breck bringing burstein cambridge christine chunks claire clark cogniac cole colin coling computational conference contributions coverage curran daniel database deerwester dependencies discourse dumais editors electronic engine environments essays evaluate evaluation evidence fellbaum finding flexible foundations from furnas geneva georgia graesser green grover harshman harter hastings high hockenmaier identification ieee indexing information intelligent interactive international james jennifer jerry jill johan journal julia kevin knight landauer language latent learning lexical linguistic linguistics marc marcu mark matheson mikheev moens morgan pages parser person precision press proceedings processing pronoun references remembering representations resolution resources science second semantic society steedman stephen steven structure student students stuff switzerland syntax systems theory tokenisation tool university using wide wiemer wordnet write
http://acl.ldc.upenn.edu/P/P06/P06-2076.pdf	222	Machine-Learning-Based Transformation of Passive Japanese Sentences into Active by Separating Training Data into Each Input Particle	aaai accuracy active addition agency aist also alternation analyzer annual asian association based basic bunrui cambridge case categorization causative chunk comparison computers conclusion conducted confirmed conll contrast conversion corpus cristianini data department dependency developed dictionaries each effective experiments feature features found haruno have heuristic high higher hirotoshi hitoshi html http hyou identification improvements index informatics information input into introduction ipal isahara japan japanese john keiko kernel kondo kudoh kurohashi kyoto language learning lexicon lower machine machines makoto manabu manually many masahiko masaki matsumoto meeting method methods most much murata nagao nara natural nello nlri numerous obtained okumura other pages paraphrasing particle particles particularly partof passive prepared press previous proceedings processing project promotion publisher publishing qing rate rates references rich rule rules sadao sato satoshi selection sentences separate separates shawe shuuei significant society software source speech springer statistical structure study support system tagging taira taku taylor technology test text thai than that these three through tinysvm traditional training transactions transformation transforming undertook university used uses using utilizing vector verb verbs version were when which yuji
http://acl.ldc.upenn.edu/P/P06/P06-1125.pdf	124	A Phonetic-Based Approach to Chinese Chat Text Normalization	according acoustics address anita annotation anomaly application approach available best between biggest blend both brown call catalog center channel character chat chen chinese cnet cocke component computational conclusions consistently constructed constructing contact corpus costs data deliver detecting different does dynamic eacl edition education effective email estimation exist experiments expressions extend final firstly from gianforte gigaword graf gram great gunter heard higher ieee ijcnlp incorporating incremental informal information institute into james jelinek journal katz kind kneser kong lafferty language learning linguistics london lrec machine made maeda mapping mappings mccullagh mercer model models modified network networks news normalization nothing november number officials online open optimal outperforms paper perform performance phone phonetic pietra pincas probabilities problem processing produces project proposed provided questions recognition recognizer references regarding remain report represent riacs rightnow rooms roossin same saunders second secondly security service sets sighan signal similarity size slash smoothing some source sparse speech stage standard statistical strings successfully technical technologies terms test text then thirdly this three time training transactions translation university varying what white with within wong words workshop xscm yuan
http://acl.ldc.upenn.edu/P/P06/P06-2021.pdf	167	Using WordNet to Automatically Deduce Relations between Words in Noun-Noun Compounds	academic algorithm artificial cognition combination combinations communication complex comprehension conceptual costello database descent devereux engineering experimental fillmore gagne hearst hierarchy identification influence intelligence investigating journal justeson katz language learning levi lexical linguistic memory miller modifiernoun natural nominals press proceedings properties psychology references relational relations review rosario selection semantics shoben some syntax technical terminology text thematic used wordnet york
http://acl.ldc.upenn.edu/P/P06/P06-1145.pdf	144	Time Period Identification of Events in Text	algorithm arthur bianca cabocha chasen class cost data dempster donald from http incomplete iterative john journal laird langford learning likelihood maximum method multi naoki proc references rubin sensitive sigkdd software taku zadrozny
http://acl.ldc.upenn.edu/P/P06/P06-2059.pdf	205	Automatic Construction of Polarity-tagged Corpus from HTML Documents	aaai acquisition added adjectives aims also although analysis andrea another answering applied applying approach automatic automatically behind bing bootstrapping both build building cicling cikm classification classifier classifiers classify concept conclusion conll consisting contextual cooccurrence corpora corpus could create creating customer dave david deal dependencies determine determining different difficult direction discuss document documents down ellen emnlp engine enhancing esuli examined excellent experiment expressions extract extracting extraction fabrizio facts first from fukushima fully future gallery general given gloss hatzivassiloglou hiroya hoffmann hong html idea identify identifying important includes information intend inui investigate jaap janyce kamps katheleen kenji kushal larger lawrence learn learned learning learns like lillian linguistic lrec maarten machine manabu many marx mckeown measure measured method mining minqing model mokken more morinaga negative nouns objective occur okumura only opinion opinions orientation orientations other ours pages pang paper parameters pattern patterns paul peanut pennock peter phrase phraselevel phrases polarity poor positive precise predicting proceedings product proposed proposes questions recognizing references relies reputations researchers result revews reviews rich rijke riloff robert rules satoshi search sebastiani seed semantic senmantic sense sentence sentences sentiment separating sets shivakumar similar similarly since some spin statistic steve subjective subsequent such summarizing synonyms tagged takamura takashi tateishi techniques terms texts than that their them then theresa these they this throush thumbs toshikazu towards turn turney unannotated unsupervised using utilized vaihyanathan vasileios wiebe wilson with wordnet words work works yamanishi yasileios
http://acl.ldc.upenn.edu/P/P06/P06-2107.pdf	253	Statistical phrase-based models for interactive computer-assisted translation	aachen adaptations advances alignment american analysis annex annual apparatus appliquee approaches april artificial assisted association august barcelona barrachina based been bender berger brown budapest canada casacuberta celer centre chapter ciones civera coling comparison completion compostela computational computer conclusions conf conference context contribution copenhagen cubel decoding della denmark department developed eacl eamt edmonton efficient emnlp empirical estimation europe european finite first formatica foster framework from gamma generation germany gillett gonzalez groups hasan have hochschule hong human hungary image improved inferred informatik informatique inteligence interactive intituto isabelle joint july june kehler khadivi koehn kong laboratory lagarda langlais language lapalme lecture lehrstul linguistics linguistique lloret lnai machine macklovitch march marcu mathematics mediated meeting mercer method methods model models monotone montreal naacl natural north notes october pages parameter patent pattern philadelphia phrase pico pietra plamondon prediction preparation probability proc proceedings processing procs project prototype rachina recherche recognition recongnition references reordering research rheinisch rwth santiago schlumbergersema science search september show single slight societe solu spain springer state states statistical step stochastic strategies structural summit syntactic system systematic target technical technische technology tecnologico templates text thesis this tomas toward transducers translation translators transtype united university used userfriendly users using various verlag vidal viii vilar volume westfalische with wong word work xerox zens
http://acl.ldc.upenn.edu/P/P06/P06-1085.pdf	84	Contextual Dependencies in Unsupervised Word Segmentation	accessor adams advances aldous algorithm analysis annals anto antoniak anyway appendix applications approximation assignment bayesian berlin bernstein between bigram brent calculated cambridge carlo categorical chain chapman chen child children chinese cohen computational corpus count counts criteria data define deng dirichlet discovery distribution distributional each easier ecole editors efficient episodes equation erlbaum estimating exactly exchange exchangeability expected explicitly extraction feng figure flour following fourth from generators gilks given goldwater griffiths hall harris hillsdale implementation information intelligent interpolating into item johnson journal kleeck language learning lingustics machine macwhinney markov meaningful mixtures model monte more must nelson neural niak nonparametric number occurring only over pages parent phonology posterior power practice preceding press probabilistically probabilites probabilities problems proceedings processes processing ratner references related replacing requires richardson saint sample sampler sampling section segmentation segmentations segmenting single snow sound speech spiegelhalter springer statistics structure substring suffolk symposium system systems table tables that then this times timeseries tokens topics track tracking types unigram using value variety volume where which with word words xiii zheng
http://acl.ldc.upenn.edu/P/P06/P06-1059.pdf	58	Improving the Scalability of Semi-Markov Conditional Random Fields for Named Entity Recognition	applied bikel biomedical christopher collier conference context daniel dingare dong entity exploiting extraction fifth finder finkel from gail gibbs grenager haechang high hybrid ijcnlp incorporating information international into introduction jenny jnlpba joint juntae kyung language learning local malvina manning method name named natural nguyen nigel nissim nymble ohta pages park performance phase proc processing ralph recognition references richard rose sampling schwartz second seonho shipra sinclair syntax systems task tateisi tomoko trond tsuruoka using weischedel yoon yoshimasa yuka
http://acl.ldc.upenn.edu/P/P06/P06-2040.pdf	186	Reduced n-gram models for English and Chinese corpora	academic accompany acoustics adaptive addisonwesley american analyse assp audio automata automatic average baayen backing backoff baker based baseline behaviour beijing belfast blasig boyle brown cambridge cancho cases categories category character china chinese city class clclp clustering cognitive coling college combination combining complex component computer computers context contexts conversational corpus darpa data department dependent design detroit digital distributions donn douglas dublin edited effort elvira english estimation eurospeech evaluation evert extension extensions ferrer framework francis frequency from function garcia george good goodman gram grammar grams hanna harald henry hermann hierarchy higher histories however human icassp icslp ieee improved improvement information integrating internationales irish island janet jianfeng jianying john joshua journ journal katz kluwer kneser kristie kucera language languages least length level lexicons liermann linguistics lnre mandarin manhung manual mari marie martin mcmahon michael midl ming model modeling modelling models modest multi natural nelson news niesler obtained origin ostendorf over owens pages paris paul perplexities peter phil philip phrase phrases present principle probabilities proc processing produces providence pruning publishers publishing quantitative queen ramon random reading recognition recognizer reduced reduction references regimes reinhard result rhode rhodes ricard ronald rosenfeld rowan rules scalable science sequences seymore seymour sicilia signal simple size slava smith smoothing snowbird sparse speech standard statistical statistique stefan stochastic street study sven table textuelles thesis thomas three topic traditional transactions trigrams turin typetoken university using utah variable varigram varigrams volume wall weighted william with woodland word words workshop zipf
http://acl.ldc.upenn.edu/P/P06/P06-2117.pdf	263	Boosting Statistical Word Alignment Using Labeled and Unlabeled Data	alignment annual association avrim basu bilenko blum brown cherry classification clustering colin collins colt combing computational conference curin data david dekang della discovery empirical entity estimation final framework franz hopkins improve international jahr john johns joint josef kevin knight knowledge labeled lafferty language learning linguistics machine mathematics meeting melamed mercer methods michael mikhail mining mitchell model models mooney named natural noah onaizan pages parameter peter pietra probabilistic probability proc processing purdy raymond references report robert semisupervised sigdat sigkdd singer smith statistical stephen sugato theory training translation university unlabeled unsupervised vincent with word workshop yarowsky yaser yoram
http://acl.ldc.upenn.edu/P/P06/P06-2095.pdf	241	Using comparable corpora to solve problems difficult for human translators	annual anthony association babych barcelona bleu bogdan computational evaluation extending frequency hartley linguistics meeting method proceedings references weightings with
http://acl.ldc.upenn.edu/P/P06/P06-2051.pdf	197	Spontaneous Speech Understanding for Robust Multi-Modal Human-Robot Communication	academy advances annals artificial automation baker bielefeld biron breazeal brooks bruce capabilities case communication companion conf cooperative dautenhahn development developmental editors extensible fillmore fiorini flexible frame fraunhofer fritsch gray haasch hagele hoffman human humanoid ieee infrastructure intelligence interactive journal kidd kleinehagenbrock language lawitzky lexical lieberman lifelong like live lockerd mulanda naccl natural nature origin other pages partners people personalized perspective prassler proc recources references robot robotics robots roman sagerer sciences semantics service speech systems text understanding verlag volume with wordnet workshop wrede york
http://acl.ldc.upenn.edu/P/P06/P06-2005.pdf	151	A Phrase-based Statistical Model for SMS Text Normalization	bangalore barzilay bilingual bootstrapping coling consensus corpus data english extracting from input instant mckeown messaging multilingual murdock normalization parallel paraphrases references riccardi summit system tochinese translation using zhang
http://acl.ldc.upenn.edu/P/P06/P06-2056.pdf	202	Unsupervised Segmentation of Chinese Text by Use of Branching Entropy	accessor analyzer ando anlp applications approach available bakeo based bell boundaries center chen chinese cleary colingacl compression computational contents context corpus crafted data deng eacl emerson engine entropy experiment experiments feng freely from hall hand harris hhmm http ictclas ijcnlp indicator international ishii japanese kanji kempe language learning lexical lexicon linguistics maximum morpheme mostly naacl natural pages part phoneme prentice references search searchable second segmentation shen shop sighan statistical tanaka text training tsou unsupervised using variety visited without witten word workj xiong yuliao zhang
http://acl.ldc.upenn.edu/P/P06/P06-2089.pdf	235	A Best-First Probabilistic Shift-Reduce Parser	american annual approach arbor association based berger best bikel briscoe budapest canada carroll chapter charniak chart coarseto collins computational corpora della design diego discriminative edge efficient engine entropy entropyinspired eugene european fine first generalised generative goldwater grammars hungary implementation johnson language large lexicalized lingual linguistics mark maxent maximum meeting michael model models montreal multi natural north pages parallelprocessing parser parsing pietra probabilistic proceedings processing references reranking seattle sharon sixth statistical three unification very with workshop
http://acl.ldc.upenn.edu/P/P06/P06-2094.pdf	240	On-Demand Information Extraction	about acquire acquisition agichtein aleksander amirault anlp antar aone appendix applied approximately aramony arikawa arimura arrest asai bank billion carlos ceridian chemicals chinatsu collec comdata company conference convict corestates corp created data database date derek discovery docid elect eslaminia eugene european event extracting extraction financial fine fleiss four from gerald gravano guilty halen hamilton havel hedayat hiroki holdings home http imprison incarcerate jail james johnson jose kane kawasone kenji knowledge kwasniewski language large last least lugo macdonald mcnally mendoza meredith merge merger midlantic mila million mitchell money month natural nguyen nist note obasanjo only optimized page part period person pkdd plaintext position potash practice president prieto princeton principles proceedings processing projects purchase queries ramos reagan rees references relation relations ronald sample santacruz scale semi sentence setsuo shinji snowball south structured substructure system tables tarango tatsuya terson that this three title upenn week wheatley wolf wordnet year years
http://acl.ldc.upenn.edu/P/P06/P06-1060.pdf	59	Factorizing Complex Mo dels: A Case Study in Mention Detection	algorithm american andrew annotated annual approaches association based benjamin british building canada cards carnegie caruana categories chapter chen chinese chunks classifier columbia combination combining communication communications computational computer conditional conference conll corpus daelemans data database dayne dempster department derivation detection douglas dynamic eacl editors edmonton emnlp empirical english entity entropy erik estimation evaluation exploring extraction factorized faculty fernando fields fien florian foundations freitag from gaussian grishman guages hacioglu hassan hladk house howtogetachinesename html human iccl icml incomplete independent index inflective information international introduction issues ittycheriah jects jing journal kadri kambhatla klein labeling laird lana language large learning lexical likelihood linguistics machine magic manning marcinkiewicz marcus markov maxent maximum mccallum meeting mellon mentions message methods meulder miles miller model models montr morphological multidimensional multilingual multitask naacl named natural ngai nicolov nist north occuring october optimization osborne pages part penn pereira plan pratt prediction presented press prior private probabilistic proceedings processing ramshaw random recognition references related report representing rich rohanimanesh ronald rosenfeld roukos royal rubin rule sang santorini schutze science segmentation segmenting sequence sequences seventh shared sixth smoothing society speech stacking stanley statistical structured sutton symbolic tagging tagset task technical technology tests text thrun tjong tracking transformation transformational treebank tutorial twentyfirst understanding university vancouver veenstra walter without wordnet workshop ying zhang
http://acl.ldc.upenn.edu/P/P06/P06-2009.pdf	155	A Pipeline Framework for Dependency Parsing	abstracted accuracy achieved acknowledgements actions activity addison additional advanced algorithm also ambiguities analysis annotated annual answering approach approaches aquaint arbor architecture arda artificial association attempting august based better between beyond bottom british building calls canada carlson classifiers coling columbia comments communication company comparing comparison compilers computation computational computer conference conll context copenhagen corpus crammer cumby current cyclic data decisions department dependency depth design designing deterministic development devise different discrete doing editors efficient eisner ellen emnlp empirical english entropy especially evaluated even existing experimentally experiments exploration featurerich following formulation framework france future generalize global globally goals good grant graph haghighi human hwee ijcai improving includes incorporate inference inferred information intelligence interestingly international into iwpt joakim joint june justify klein labeling language large learning level linear linguistics local long machines made making manage manning marciniak marcinkiewicz marcus margin matching matsumoto maximum mcdonald meeting methods michigan minimizes model models more naacl nancy national natural necessity network ninth nivre number observed october online only optimization other pages parser parsers parsing part penn pereira perhaps pipeline principles private probabilistic proc proceedings process processing produced program programming projective providing publishing punyakanok question ratnaparkhi reading references reflex relative relatively reliable report research resolve results riloff robust role rosen roth ryan santorini scholz science second semantic sentence sentences sethi several shallow show showing significantly snow sources speech statistical strube study such suggestions suggests support supported syntactic system systems table tagging tasks technical techniques technology text textual than thank that these this those three time tools toutanova training treebank trees trying uiuc uiucdcs ullman under unified used useful vancouver vasin vector well were wesley when where which with work yamada
http://acl.ldc.upenn.edu/P/P06/P06-2104.pdf	250	A Comparison of Alternative Parse Tree Paths for Labeling Semantic Roles	accurate alternations annotated annual association automatic baker based beatrice berkeley beth building charles charniak chicago christopher classes coling collin collins computational conference corpus daniel dekang dependency driven english entropyinspired eugene evaluation fillmore first framenet gildea granada head international investigation john jurafsky klein labeling language large levin linguistics lowe manning marcinkiewicz marcus mary maximum meeting michael minipar mitchell models montreal naacl natural parser parsing penn pennsylvania preliminary press proceedings project references resources roles santorini semantic spain statistical systems thesis treebank university unlexicalized verb workshop
http://acl.ldc.upenn.edu/P/P06/P06-1147.pdf	146	Utilizing Co-Occurrence of Answers in Question Answering	abductive answer answering approach bakshi bensley bowden carrol chua chung clark combining context czuba eacl extraction ferrucci gerber harabagiu hermjakob hovy huynh improve junk karger katz korea language main mining moldovan multi multisource national natural prager precision proceedings processing quan question reasoning references relations role selectively singapore sinha song strategy system systems task techniques trec university using webclopedia welty williams with workshop
http://acl.ldc.upenn.edu/P/P06/P06-1087.pdf	86	Noun Phrase Chunking in Hebrew Influence of Lexical and Morphological Features	abney adler based berwick carol chunks coling disambiguation elhadad hebrew meni michael morpheme morphological parsing proc references robert sidney steven tenny unsupervised
http://acl.ldc.upenn.edu/P/P06/P06-2047.pdf	193	Graph Branch Algorithm: An Optimum Tree Search Method for Scored Dependency Graph with Arc Co-occurrence Constraints	algorithm algorithms analyser analysis annual appear arborescence association based bestfirst bounding branch branchings bringing bureau capacity choi cognition coling colingacl combinatorial computer constraint control convention corpora data dependency directed edmonds ehara eisner engineering evaluation exploration fast fifth forest generative grammar grammars graph hajic harada harper hirakawa hltemnlp ibaraki incremental information japanese journal jsai kanahe katoh language large maruyama mcdonald measures meeting method mizuno models nasr national natural nilsson nivre nonprojective optimization optimum ozeki packed pacling pages parsable parser parsing pereira polynomially preference probabilistic probablistic problems procedure proceeding proceedings processing projective projectivity pseudo rambow reestimation references representation research ribarov sage science sciences search semantic shared shortest sinica society software space spanning standards state statistical structure system three together transactions tree using very wang weak workshop zhang
http://acl.ldc.upenn.edu/P/P06/P06-3006.pdf	277	Semantic Discourse Segmentation and Labeling for Route Instructions	advances algorithms another available bartlett chasen classification collins conditional conference data directions exponentiated fields follows gradient haas http information international labeling lafferty largemargin learning machine mcallester mccallum models neural nips pereira probabilistic proceedings processing random references robot segmenting sequence simulated software structured systems taku taskar testing that toolkit unpublished
http://acl.ldc.upenn.edu/P/P06/P06-2099.pdf	245	Compiling a Lexicon of Cooking Actions for Animation Generation	adachi based between cooking couple definitions generation hisahiro into language make method natural note order pacific pages proceedings processing recipes references required salt sand shellfish similarity spew symposium that water
http://acl.ldc.upenn.edu/P/P06/P06-2088.pdf	234	Simultaneous English-Japanese Spoken Language Translation Based on Incremental Dependency Parsing and Transfer	algorithms analysis ando annotated annual arak arakocsicnarf architectures association based blackk brecher brow building campbell casacuberta cnarf collins communication comparison computational conference consecutive construction conversation corpus cross denver development driven eertt engineering english erutcurts eruttcurtts esenapaj esrap evaluation example features field figure finite flow francisco frederking from furuse hanazawa head hoge hsiillgne ibot iboyusteg ieee ieice iiatt iibott iida iiomo improving inagaki incremental information international interpretation interpreting iomo ishikawa isotani japan japanese jbus journal kawaguchi knowledge kurohashi language large lingual linguistics lliin lrec machin machine make marcinkiewicz marcus matrix matsubara matsubrara mediated meeting mima models monday moody morf morff morimoto nagao natural next noise ocsicnarf ocsiicnarf ohara ohno pacific pages parallel parsed parsing pdas penn pennsylvania picheny proceedings processing project proposal real references report research resources revned robustness sagisaka santorini simultaneous software speech spiral spoken star state statistical stein studies sugaya symposium syntactically system systems takezawa tech temporal testing thesis tnaw tongues transactions transfer translation travel treebank ttnaw ttupnii ttupttuo txen uhsiar university usam using utilizing vidal vilar want while workshop yadnom yamada yamamoto ycnedneped yllff yokoo
http://acl.ldc.upenn.edu/P/P06/P06-2116.pdf	262	A Grammatical Approach to Understanding Textual Tables using Two-Dimensional SCFGs	abraham abstraction advanced agha algorithm andrea annual answering appro approach association august bouayad canada case cfgs chaudhuri cikm classifying computational computer concepts conclusion conference context costagliola data database dayal diagrammatic dimensional discovery document donia edinburgh editing edition efficient elements estimation extraction feder find finer formatting framework francis fraser free gennaro goodman grammars harvard have hawaii henry hill html hurst information inside integrate interfaces international internet interpretation introduced invited jerome jiawei joshua kamber kaufmann knowledge korth language languages lari layout linguistically linguistics lucia management march master matthew mcgraw meeting method micheline mining more morgan nadjet ninth olap ontario orefice oriented outside overview pages parsing plex power press proceedings queries record references richard sciences scott search semantic sergio shih shun sigmod silberschatz simon speech stochastic structure sudarshan sung support surajit system table tables tabular talk techniques technology texts thesis torisawa towards tsujii umesh university using visual viterbi wang warehousing waterloo wide widm with workshop world xinxin yang yingchen york yoshida young
http://acl.ldc.upenn.edu/P/P06/P06-1001.pdf	0	Combination of Arabic Preprocessing Schemes for Statistical Machine Translation	agbago alignment american americas amta analysis analyzer annotated applied approach arabic arbor association automatic automatique bangalore barcelona base based beam better bies bleu bordel boston brooklyn buckwalter building burch cairo callison canada catalog chapter chunks computational computing conference consensus consortium corpus data decoder description diab division eacl egypt emnlp empirical engine enhanced error european eval evaluating evaluation explicit extensible fell foster frederking from generation germany goldwater google guided habash hacioglu heads heights hwang hybrid hypotheses ieee imamura improving information international italy iwslt japan jayaraman johnson jurafsky koehn kuhn langage language languages large lavie lexeme linguistic linguistics lisbon lrec maamouri machine martin matching matusov mcclosky method methods michigan minimum model modeling models morocco morphological morphosyntactic multi multiple naacl natural naturel nemlar nirenburg nist nobody nomoto north okuma osborne papineni parallel part paul penn pennsylvania perfect pharaoh phrase phrasebased popovic portage portugal preprocessing proc proceedings processing rambow rate recognition references report research resources riccardi role roukos sadat sapporo scale scarce schemes search significance spain speech spoken srilm statistical stems stolcke stuttgart suffixes sumita swoop system systems tagging talk taln technical tests text texts than three through tikuisis tokenization toolkit tools towards training traitement translation treebank trento ueffing understanding university unpublished using vancouver version voted ward with word workshop york yorktown
http://acl.ldc.upenn.edu/P/P06/P06-4005.pdf	291	An intelligent search engine and GUI-based efficient MEDLINE search tool based on deep syntactic parsing	based bioinformatics blaschke chun dictionaries discovering disease domain extract extraction frame from gene hishiki huang ieee information intelligent interactions learning literature machine medline module nagata pages part patterns proc protein references relations shiba suiseki system systems tsujii tsuruoka using valencia
http://acl.ldc.upenn.edu/P/P06/P06-2020.pdf	166	Topic-Focused Multi-document Summarization Using an Approximate Oracle Score	agreement among analysis annotation automatic back barcelona basciss based basics bleu brain carbonnell center classy college conference conroy content copeck dang daniel decomposition dianne diversity division document documents dunlavy ellen emnlp evaluating evaluation factoid frequency goldstein goodman halteren hans hidden http human impact infogistics information jade jamie jing john judith kishore leary leftbrain lucy machine march markov mary maryland matrix method microsoft model models multi nenkova nist nlprocessor okurowski overview pages papineni park performance pivoted posdemo proc proceedings producing references reordering report reranking research right roukos salim sarah schlesinger sigir simone source stability stage stan summaries summarization system szpakowicz tasked technical terry teufel text thomas three todd trang translation understanding university vanderwende vocabulary ward watson workshop
http://acl.ldc.upenn.edu/P/P06/P06-1110.pdf	109	Advances in Discriminative Parsing	abeille abney accuracy acknowledgments adaboost advances algorithm also anonymous apply approach artificial atomic authors avoiding based baseline best bikel black boosting both bregman building case challenges chap charniak chris clark classification classifier cleverness coarse collins comments comparing complexity component compound computational computationally conclusion confidence constituent constraints constructive contains corpora coverage criticism crouch curran cynthia decision deep dependency descent design discriminative dissertation distances doctoral does driven during eacl efficient effort eisner emnlp engineer engineering english entropy experiments exponential fast feature features fewer fine first flickenger function future gdaniec generate generative give gradient grafting grammars grants grishman hard harrison head helpful henderson hope hypothesized icml implementation improve improved improvements incremental incrementally infer inference inferences information intelligence intricacies invariance irrelevant isozaki iwpt johnson joint journal kaplan king klein know koller kudo lacker language lastly lavie lead learning length like linear linguistic linguistically linguistics logistic machine made magerman make manning marcus margin maxent maximum maxwell melamed merely might minimal model models modern more naacl natural network neural norvig other over overfitting overview parse parsed parser parsing part penn perceptron perform performs perkins pike plan powerset prediction predictions present problems procedure processing quantitatively rated ratnaparkhi reduce references regression regularization required requisite reranking research resistant reviewers riezler roark rotational rudin russell sagae santorini schapire selection shall shallow shift shorter should singer smith soft some sophisticated space speech speed sponsored standard statistical stochastic strategy structured subtree such suggest surpasses suzuki syntactic syntax tagger tags taskar tasks taylor thank that theiler their then this time times training translation tree treebank treebanks turian using vasserman well wellington whose will with without work workshop would
http://acl.ldc.upenn.edu/P/P06/P06-2070.pdf	216	Stochastic Iterative Alignment for Machine Translation Evaluation	alberto alex alignment alon american annual approach arbor association automatic banerjee bannard barzilay bilingual blatz burch callison chapter chris colin computational conference confidence corpora correlation cyril erin estimation evaluation extrinsic fitzgerald foster gandrabur george goutte human improved intrinsic john judegments kulesza lavie learning lillian linguistics machine measures meeting meteor metric michigan multiple naacl nicola north pages parallel paraphrase paraphrasing proceedings references regina report sanchis satanjeev sequence simona summarization technical translation ueffing unsupervised using with workshop
http://acl.ldc.upenn.edu/P/P06/P06-2041.pdf	187	Discriminative Classifiers for Deterministic Dependency Parsing	accuracy acknowledgements across adwait algorithm alon american analysis analyzer annotating annotation annual antal arpa asahara association bart based beatrice beskriv best better bies black bosch britta cambridge canon cascaded chang chapter charniak cheng chih chinese chiou christopher chung chunking class classification classifier close coarseto coling collins combined comparable comparison complex complexity computational computing conclusion conference conll corpus council coupling crammer daelemans daphne darpa data david decision department dependency deterministic ding discriminative divided dong driven duffy ecml effects efficiency efficient einarsson emnlp empirical engineering english entropy entropyinspired estimates estimators eugene european evaluation even examining ezra feature features ferguson fernando fien finder fine first fourth frederick from geman generative gives global grammars grammatisk grateful hall have head hierarchical higher hiroyasu history hoste human iccc interaction international into involved iwpt japanese jelinek jennifer jens joakim johan john johnson journal karen katz kenji klein koby koller kudo lafferty language languages large lavie learning lexicalised lexicalized library libsvm linear linguistics lrec lund machine machines macintyre magerman manning manual marcinkiewicz marcus margin mario mark martha mary masayuki matsumoto maxent maximum maxmargin mcdonald meeting memory mercer method methods meulder michael mitchell models multi naacl natural nature naudts nianwen nigel nilsson ning nivre node north observed obtained online optimization pages pairwise palmer paper parameter parser parsers parsing partially penn pennsylvania pereira performed phrase predicateargument presented press probabilistic probability proceeding proceedings processing projective ratnaparkhi references reranking research resources respect richer riezler robert root roukos ruby ryan sagae salim santorini scandinavian schasberger scholz second selection series sets shows sighan sixth skriftsprakskonkordans skriven smaller sons speech springer state statistical stefan steven stochastic structure stuart studentlitteratur superior support supported svenska swedish taku talad talbankens taskar tasks technologies technology teleman text that theory thesis third this three time ting towards training tree treebank unification university using vapnik varying vector veronique vladimir volkan vural walter weng whereas wiley with work workshop yamada york yuan yuchang yuji zhiyi
http://acl.ldc.upenn.edu/P/P06/P06-2122.pdf	268	Inducing Word Alignments with Bilexical Synchronous Trees	albert alshawi annual arbor association bangalore based bilexical chen chiang cliffs collections compiling computational conference context cruz david dependency douglas efficient eisner empirical englewood finite free giorgio goodman grammars hall head hierarchical hiyan jason jeffery joshua language learning linguistics machine michigan model modeling models pages parsing phrase prentice proceedings references santa satta shona smoothing srinivas stanley state statistical study techniques theory transducers translation ullman volume
http://acl.ldc.upenn.edu/P/P06/P06-4015.pdf	301	The S A M M I E System: Multimodal In-Car Dialogue	about academic acapulco acceptance access actual adaptive albums alexandersson allen almost also annual application approach april architecture association august baldridge based baseline basic becker between blaylock budapest bunt categorial change chapter collaborative collect collecting collection combinatory completed computational conducted conference context coordination could current data database david decide details dialog dialogue directions discourse distraction driver driving dybkj eacl editors enlg european evaluation experiment experiments explore freedom freely fusion gave generic gerstenberger grammar half haptic have http hungary icmi igfa ijcai implementation information input intelligent interaction interactive interfaces international involved just kipp kluwer knowledge korbayova kruijff laila lane language large larsson linguistics lisbon management mattes maybury meeting million minker modal model mono more multi multimodal multiple music only openccg operation order output overlay pages perform performance pfleger planning player poller practical presentation press primary problem proc proceedings processing realization reasoning references response rieser roadmap robust schehl seattle semantics september session setup sigdial simulating simulator solving songs sourceforge speech spoken staffan state stock strategies subjects system systems task technology text than their them tool tools traum usability used user using variety version volume wahlster washington with wizard wizards wolfgang workshop york zancanaro
http://acl.ldc.upenn.edu/P/P06/P06-4013.pdf	299	Archivus: A multimodal system for multimedia meeting browsing and retrieval	accessing agnes ahrenberg algorithms analysis appli applying archivus arne bengio berlin bourlard cation cenek computer conference content coutaz czech dahlback daniel design dialogue dianne domain editor editors ewhci framework from geneva gray hefley humancomputer indications intelligent interaction interface interfaces international joelle joint jonsson june karlovy lars learning lecture lisowska lncs london machine martigny martin matousek mautner meeting meetings melichar miroslav multimodal murray nils notes november pages papers pascal pavel pavelka preliminary press procedings proceedings project query rajman rapid recorded references related report republic salber science selected september speech springer studies study switzerland system systems technique text third tomas trung university user vaclav vary verlag volume wayne william wizard workshop
http://acl.ldc.upenn.edu/P/P06/P06-2106.pdf	252	Infrastructure for standardization of Asian language resources	afnlp bertagna calzolari content http interoperability issues language lenci lexical mile monachini open perspectives references resource resources
http://acl.ldc.upenn.edu/P/P06/P06-1019.pdf	18	Partially Specified Signatures: a Vehicle for Grammar Modularity	based bender broad candito coling computation consistent copenhagen coverage cross denemark development emily engineering flickinger fouvry fredrik grammar grammars helene hierarchical language linguistically ltags marie matrix melanie multilingual oepen open pages precision principle proceedings rapid references representation research shared siegel source starterkit stephan taipei taiwan workshop
http://acl.ldc.upenn.edu/P/P06/P06-1035.pdf	34	Measuring Language Divergence by Intra-Lexical Comparison	acoustic activation adaptive against akademii algorithms america american analysis anatolian andean april atkinson auditory bailey basis basseville benedetto benefits between binary black borrowing both boyd brett brew caglioti called cambridge capable case chapter charles chris cladistics classification codes cognitive compares comparison complex computational conclusions conference constructs contacts correcting cristianini dataset dating december deletions determinants dialect directions discrimination disfluency distance distancebased distances divergence doklady duration dyen each edinburgh edition editors estimation ethnic ethnologue european evans experiment exploring families filled finding fisher form formal from future gary generalization genetic goldinger gray grimes group gruyter grzegorz hahn have hearing heeringa heggarty indiana indo indoeuropean inference insertions interest internally international invariant isidore jeffreys john joseph journal kapatsinski kernel kessler kidd kirby kondrak kruskal language languages leary letters levenstein lexica lexical lexico lexicon lexicostatistical london loreto luay luce lyle mcdonald mcmahon measure measures measuring meeting memory mental method methods micchelli michle mitchell model modeling models monica mouton multivariate nakleh natalia nature nauk neighborhood neighborhoods nello nerbonne network noakes oliver only origin pages paper pattern patterns paul pauses perception perspectives philological philosophical phonetic phonetically phonology phonotactics phylogenetic physical pisoni populations port prehistoric presented press prior probability problems proc proceedings processing proportion psycholinguistic psychologically quantitative quentin real recognition recognizing reconstruction references relations reliable report research reversals review richard ringe robert rule russell sankhya science scientific scott seems shawe shepard shillcock signal sigphon similarity simon slaska society sound special speech spoken spontaneous sssr statistic statistical status steven structure study sublists support swadesh tamariz tandy taylor technical that their theory thesis this times todd toronto total toward transactions tree trees ulrike universal university variation vsevolod warnow watson where wordlikeness words world zipping
http://acl.ldc.upenn.edu/P/P06/P06-2027.pdf	173	Automatic Creation of Domain Templates	aaai acquisition activities algorithm alignment alvin amit analyzing andrew annotated answering approach arikawa arimura asai atomic automatic bagga baker bank barzilay bikel biographical biographies biography blair boyan broadcast bruce cardie case charles christopher claire coling collier collin complexity computational conceptual content corpora corpus creation daniel darpa data david definitional dennis design detection development discovery document doddington domain dragomir duboue eduard efficiently elaine elena ellen emnlp empirical encoding evaluation event events exact extraction factored fast filatova fillmore finley fiscus fleiss forest frame frequent from garofolo generating generation george gerard gildea gistexter goldensohn grishman hacioglu hans harabagiu hatzivassiloglou hazen hiroki hobbs hovy huttunen hybrid improved independent inference information israel issue issues james jerry jinxi john joseph journal jurafsky kadri kathleen kawasoe kenji kingsbury kiri kiyoshi klein korelsky labeling lacatusu language large learning learns levin lexical liang licuanan lillian line linguistics literary lrec luhn machine machines maiorano manning mark marsh martha martin mckeown mechanized methodology methods michael mining miruna model mohammed multi multidocument multiple multiplesequence myunghee naacl name natural news nips nist occupation onyshkevych optimized other overview pablo pages paik palmer paraphrase parsing pattern paul perzanowski pierce pkdd pradhan prager prenticehall press principles proc proportions proposition questions radev ralph ranlp rates references regina related representation research resources respect results retrieval richard riloff robin roles roman rules salton sameer sanda sasha satoshi scenarios schlaikjer schmelzenbach schwartz searching sekine selection semantic semantics semi setsuo shallow sheffield shinji sigir sigkdd silja smart sources special statistical steven structure structured substructure sudo summaries summarization support system tanya task tatsuya technology tell template text that themes thesis ticrea topic tracking trees understanding university unsupervised using vasileios vector very vincent wagstaff ward wayne weischedel what white wiley with wordnet workshop yangarber zaki zhou
http://acl.ldc.upenn.edu/P/P06/P06-2030.pdf	176	Using Bilingual Comparable Corpora and Semi-supervised Clustering for Topic Tracking	acquisition addressed alarm alignment allan also analysis application approach article background banerjee basu because better between bilingual calculating carbonell categorization centroid chasen chen choosing church classifying clustering clusters coling collier comparable compared comparing comparison conclusion constrained coordinating corpora correlation cosine dagan darpa defined detection dictionary difference does each eacl effect efficient empirical encouraging english erroneously estimation event examine existing extending extracting false figure fixed found franz further furthermore future german goodman high hirakawa humans icml illustrates improvements improving includes incremental indicates information investigated issue japanese journal judged knowledge kumano language large larkey lavrenko line machine machines manual matsumoto maximum mccarley means method methods minimum miss model modeling mooney moore morphological multilingual naist nallapti need negative news number oard obtained ones other pairs paper papka part pelleg perform positive predefined prise probability proc rate references report result results retrieval schmid seeding segmentation semi semisupervised sigdat sigir similarities size sizes smoothing specific speech stories story study such supervised surprising system tagging task technical techniques term termight terminology test text than that there therefore this topic topics tracking training translation umass unsupervised using wagstaff well were with without work workshop yang
http://acl.ldc.upenn.edu/P/P06/P06-2015.pdf	161	An Account for Compound Prepositions in Farsi	abdolazim abdorahim account acknowledgement advanced affixes ahmad algorithms aliashraf also although amirkabir analysis angeles applicable application area argued asrafi atro automatic baker bateni been believe believes better bonyad building bzsazi bzsensie california categories category challenging changing chicago classbased compound computational contributes cultural dastare dasture dasturie dasturnme datadriven deconstructing department different disambiguating discipline diss document education ehyye elmi elmie emruz encouragements english eslami esteqqie ezafe ezfe farhangy farokh farshidvard farsi farsie ferdowsi figure finding fixed formation foundations fresno from frsi frsie function functional gastri gharib ghayoomi gholm goftar goftre grammatical grams hall handbook harfe hausser help helpful helping here higher homayoun horufe humanities hypothesis important incorporation information institute introduction iran iranian javad journal jurafski jurafsky kalameye kalbasi karborde ketabe ketb khalil khanlari khatibrahbar khosrow knowing labeling language languages least lexical lieber like linguistic linguistics literature literture mabanie martin mashad mashkur masood mehdi meshkatodini meyre mitkov moaser models mohammadreza moharam morakab morphological morphology motivated nahve natural navye nazariye neutralization nouns novin ones order ostad other output oxford panj paper parts parviz payeye pearson persian phrasal phrase poor prentice preposition prepositional prepositions press proceedings processing publication rabt rather rayaneie recognition references research researches retrieval roland ruslan sadegi sadi samiian sarf seem select seminar senaxte sense seyed shafaii shargh since sort special speech springer state stemming structure studies such suited supports sxte sxtemane syntactic syntax taggers take tasxise tehran tell than thanks that theory this tosife university used vafai very vida voutilainen ways which word words zabne zade
http://acl.ldc.upenn.edu/P/P06/P06-1093.pdf	92	Automatic Generation of Domain Models for Call Centers from Noisy Transcriptions	accent accurate acknowledgements acoustics acquisition agarwal algorithm algorithms allocate allows alonso already also analysis april areas audio automated automatic automatically available balakrishnan barcelona based bechet bell bremen briefly browsing build call calls calltype captures care carmel catalogs center classification cluster clustering colleagues complex comprehensive conference conversations corpora corpus customer daily data december description dialogs dialogue different difficult discriminative discussion discussions document documents domain domains douglas effect effectively eliminated emnlp empirical eurospeech evaluation experiments extracted foreign form fragments framework from future generation geneva germany gilbert given gorin grammar greece grieco hacioglu haffner hakkani handles harris have helpful hierarchical hong hoory huang hundred identify ieee improve information international islands jiang july kingsbury knowledge kong krishnapuram kummamuru language large lawson learnt like link list lotlikar louisiana management mangu manual manuals methods mining mishne model modeling monitor monothetic more nato natural news noisy november october olivier ontological optimizing orleans padmanabhan paper part pellom performance possible presented problem processing provides providing quickly raghuram recognition redundant references research resources results retrieve rhodes riccardi routers roytman salient saon scenarios search semantically september several showed shown signal singal siohan soffer spain specific speech spoken sreeram succinctly suggested summarization svms swayne switzerland system tang task taxonomies taxonomy technical text thank that this thomas thousand through topic topics training trans transcription transcriptions type understanding unsupervised updated used useful various very view virgin voice voicemail volinsky watson where wide with work workshop world would wright year york
http://acl.ldc.upenn.edu/P/P06/P06-1091.pdf	90	A Discriminative Global Training Algorithm for Statistical MT	acknowledgment alberta alexandria algorithms aligner anonymous anoop arabic arbor authors automatic based bleu boston british brown byrne canada catlett chiang christoph collins columbia conference contract crammer criticism daniel darpa david della dependency detailed discriminative edmonton eleventh emnlp english entropy error estimation evaluation experiments fernando franz gale heterogeneous hidden hierarchical international ittycheriah japan joint josef july june kishore koby koehn kumar large learning lewis libin like local localized machine main marcu margin markov mathematics maximum mcdonald mercer method methods michael michigan minimum model models naacl nist niyu october onaizan online pages paper papineni parameter parsers partially perceptron pereira peter philadelphia philipp phrase pietra prediction proc proceedings project rate references reordering report reranking reviewers robert roukos ryan salim sampling sapporo sarkar shankar shen site statistical stephen supervised supported thank their theory this tillmann todd tong training translation uncertainty under vancouver vincent ward weijing william with word work workshop would yaser young zhang
http://acl.ldc.upenn.edu/P/P06/P06-2004.pdf	150	The Effect of Corpus Size in Combining Sup ervised and Unsup ervised Training for Disambiguation	aaai adam alexander ambiguity association attachment atterer backed backoff based bikel brooks calvo charniak cicling collins comparison computational conll context corpora daniel distributional donald enhancing eugene framework free from gelbukh generative grammar hindle hinrich hiram iaai information intricacies james kilgarriff large lattice lexical lexicalised linguistics mats michael michaela model models pages parsers parsing prepositional references relations rooth schutze statistical statistics structural techniques thesaurus third three through unlabeled unsupervised very with word wordnet workshop
http://acl.ldc.upenn.edu/P/P06/P06-1075.pdf	74	The Effect of Translation Quality in MT-Based Cross-Language Information Retrieval	across akira amano annie annual approach automatic based bilingual cambridge chen christian clir cole collier computing conference cooccurrence corpora cross david demner development dictionary dina doddington douglas effect englishjapanese evaluation exploring fluhr franz fushman future gareth george gram grefenstette gregory hans hawaii hicss hideki hirakawa hirosysu hsin hull human hyon information infromation international jing jones joseph kando kazuaki kazuko kazuo kishida kuang kumano kuriyama language languages list machine mariani martin mccarley multilingual myaeng nigel nogami noriko oard overview pages parallel press proc quality quantifying querying references research resources retrieval ronald sakai sciences scott second shin sigir size state statistics sukhoon sumita summit sung survey system technology term tetsuya todd toshiba translation university using uszkoreit utility victor ward workshop york zaenen
http://acl.ldc.upenn.edu/P/P06/P06-1088.pdf	87	Multi-Tagging for Lexicalized-Grammar Parsing	accurate adcock algorithms almost analyser analysis annotation anthony approach aravind artificial bangalore barcelona briscoe budapest canaria carnegie carroll cassandra categorial charniak chen clark coling collins combinatory computational conference coverage curran della dependency discriminative eacl emnlp engineering entropy eugene experiments features fields gaussian general generative geneva glenn gotoh grammar gran hidden hockenmaier hungary ieee importance inducing intelligence investigating italy james jeremy john joshi julia katz lafferty language lightweight linear linguistics littman lrec machine mark markov maximum mccann meeting mellon methods michael models natural pages palmas parsers parsing partial pattern perceptron philadelphia pietra pittsburgh prior proceedings random references report robust ronald rosenfeld smoothing spain srinivas stanley statistical steedman stephen supertagger supertagging switzerland taggers technical theory training transactions university using venice vincent wide with workshop yoshihiko
http://acl.ldc.upenn.edu/P/P06/P06-2086.pdf	232	URES : an Unsupervised Web Relation Extraction System	aaai achieve acquisition addition agichtein allowed analysis anlp applying approach arlington articles artificial attributes automatic automatically autonomously background bikel biomedical bootstrapping bottleneck brin broadcast bypasses cacm cafarella califf cambridge chinchor cikm classic collections collins complex complexity computational conclusions conf conference constants context contexts corpora corpus created currently darpa database description developed dictionaries digital discourse downloaded edbt either elements emnlp empirical english entity entropy etzioni evaluating experimental exploiting explorarions explore extend extracting extraction feldman finder fisher foundations francisco free freitag from further future generating grammar gravano handle have herndon heuristics higher highperformance hirschman human hybrid iaai inductive information instead intelligence international japanese jones just kaufmann lack language large learn learning level lexical lexicons libraries like linguistics logic machine manning manually markov match maximum mccallum menlo message method methods miller models mooney more morgan multi name named national natural news notes nymble overview park parse pattern patterns phillips plain precision predicates presented press probabilistic proc proceedings processing programming queries references relational relations relies riloff rosenfeld rules schutze schwartz segmentation sekine semantic semi sentences simple sixteenth sixth skips slots snowball soderland some spring statistical strong structured study sudo symposium syntactic system systems tagged tagging task technology test text that thelen third topics training types umass understanding unlabeled unsupervised untagged upon ures used using very want webdb wide with wordnet work working workshop world would zelle
http://acl.ldc.upenn.edu/P/P06/P06-1073.pdf	72	Maximum Entropy Based Restoration of Arabic Diacritics	abdou achour afify analyzer approach arabe arabic audio automatique berger buckwalter chen computational computer consortium contemporain correspondances data debili della empirical entropy etiquetage goodman grammatical ieee institut isbn joshua language linguistic linguistics maghreb makhoul maximum modeling models morphological natural nguyen palisades pietra processing recherche references report ronald rosenfeld smoothing souissi speech stanley study survey system technical techniques trans version voyellation workshop xiang
http://acl.ldc.upenn.edu/P/P06/P06-2023.pdf	169	A Bio-inspired Approach for Multi-Word Expression Extraction	advances algorithms aligning alignment annual applications approach arabic arantza argamon association baldwin banerjee based basenp beatrice bigram binnenpoorte bioinformatics biology bond boves burkhard case casillas catia changning character collections collocations common composition computational computer conference contiguity copestake cucchiarini dagan daille daniel data dependency design diaconescu diana discrete documentation endong english eric exact experimental exploratory expression expressions extraction flickinger francis frequency generative gram grammar grams hans helmer hirschberg huang hybrid identification ifip implementation information intelligent international ivan journal knut krymolowski language learning lecture lenhof linguistics longest luka management martlnez matching mathematics meeting memory mielikainen ming mining model molecular morgenstern multi multiple multiword mustafa natural neck nerima ngram notes package pain pattern patterns pedersen peter problem processing pronunciation raquel recent references reinert robertson satanjeev science searching segment segmentto sequence seretan shallow shlomo siam smith solution speech spoken statistical statistics stefan strik study subsequence subsequences suleiman syntactic systems taneli terminology terms text textual theoretical timothy translation unified using variation views violeta waterman wehrl willett word workshop yuval zhou
http://acl.ldc.upenn.edu/P/P06/P06-1047.pdf	46	Extractive Summarization using Inter- and Intra- Event Relevance	aaai achieves algorithm allison analysis approach approaches articles artificial arul associate associations automatic average banko barzilay based below best both brin bring centrality centric chin chosen christiane chunfa cicling citation closely coherence concept concepts conclusion consider constructed containing contextual cooccurrence daniel data database derive derived deriving document dragomir each eduard electronic elena elhadad entities entity erkan evaluate evaluation event eventbased extractive fellbaum filatova finding formal from generation gram graph graphs grobelnik gunes haraguchi hatzivassiloglou highest hltnaacl hovy html http important improve including independent information integrate intelligence inter interested intra investigated issue jason journal jure language lawrence learning leskovec levels lexical lexrank linkkdd local long lucy making makoto marko masaharu measuring menezes michael michele michelizzi mihalcea milicfrayling mingli modelling more motwani multi multiple naacl named naomi natasa news nist nlpir notes ntcir ontology order outperforms page pagerank paper patwardhan pedersen poster press proceedings proceeings projects propose rada radev rajeev ranking reference references regina related relatedness relevance report reported representation research result salience same scores semantic sentences sergey siddharth significance significantly similarity stanford statistics structures summaries summarization summary syntactic technical terms terry text that they this timothy tokyo university upenn used using vanderwende vasileios wenjie whose winograd with wordnet work working workshop yoshioka yuan
http://acl.ldc.upenn.edu/P/P06/P06-1048.pdf	47	Models for Sentence Compression: A Comparison across Domains, Training Requirements and Evaluation Measures	aaai accurate akira ambiguity anlp annie annotation approach arbor artificial audio automated automatic bangalore barcelona based beyond blind brian briscoe british burnard cambridge canada carroll casimir charniak chiori clarkson classification coling compaction compression computer computing condensation consortium constraints corpus corston crouch daniel disambiguation discriminative display eacl edition edmonton entropy eugene eurospeech evaluation example expert extraction flannery francisco from fukushi functional furui general generation geneva grammar greece grefenstette gregory guide hidden hongyan hori horiguchi hybrid ieice information inlg inspired intelligence intelligent israel italy jenine jing kaufman kaufmann kevin king knight kulikowski language learn learning lexical lrec machine machines marcu markov masaru maximum mcdonald method methods metrics minh mitzpe model modeling morgan naacl national nets neural nguyen numerical oliver owen oxford packing pages palmas parser philip pittsburgh prediction press probabilistic proceedings producing programs provide publishers quinlan rambow ramon recipes reduction reference references rhodes richard riezler robust ronald rosenfeld ryan sadaoki saul scanning scientific screens seattle sentence series service shimazu sholom simon small soft spain speech srinivas stanford statisti statistical statistics stefan steve stochastic subtitling summarization supervised support susumu switzerland symposium syntactic systems talip telegraphic teukolsky text that through toolkit tracy transactions trento turner university unsupervised users using vandeghinste vector very vetterling vincent weiss whittaker william with word workshop world york zaenen
http://acl.ldc.upenn.edu/P/P06/P06-2033.pdf	179	Conceptual Coherence in the Generation of Referring Expressions	aloni annual association barkerplummer barry beaver benthem church cognitive computational cooking cover csli dale definite descriptions diagrams editors ellis experimental expressions gardent generating generation gricean hanks information interpretation journal lexicography linguistics luzio maxims meeting minimal morrison mutual naming norms pictures proc proofs psychology quarterly questions references referring reiter robert science scotto snodgrass stanford under vanderwart word words
http://acl.ldc.upenn.edu/P/P06/P06-1049.pdf	48	A Bottom-up Approach to Sentence Ordering for Multi-document Summarization	aaai agreement algorithms also although among analysis annual anova appeared applications approach arrange arranging artificial association automatic barzilay becomes better bleu bottom catching chichester chronological coefficient coherence coling compare compared computational conclusion conference content continuity continuous corpora could criteria daniel decrease dependencies described difference differences different direction directly document dragomir drift effects eleazar elhadad eskin evaluation even existing experiment experimental experiments extracted figure four from functional future generating generation global hatzivassiloglou higher honest hovy however iaai implied improvement improving increases inferring intelligence international ishizuka island journal judith kathleen kathy kendall kishore klavans language lapata learning lemmatized length lillian line linguistics local lower machine main major mann marcu matsuo mckeown measuring meeting method methods metrics mirella mitsuru models multi multidocument multiple naacl naoaki national natural neats news noemie noun nouns number observed okazaki ordering orderings organization over pages papineni performed planning play played precedence precision present probabilistic proceedings produce progress prospects providence radev references reformulation regina relation reported research rest results revealed rhetorical rhode role roukos salim section segments sentence sentences showed shows significant significantly some sources sparse statistical strategies structure structuring summaries summarization summarizer table test text than that theory therefore these this thompson though todd toward towards translation tukey understanding unit value values vapnik variance vasileios verb verified ward weijing wiley with workshop yutaka
http://acl.ldc.upenn.edu/P/P06/P06-1043.pdf	42	Reranking and Self-Training for Parser Adaptation	adaptation adrian andrew artificial assoc bacchiani best biomedical brian cambridge charniak clegg coarseto cohen computational computer corpus discriminative empirical eugene evaluating fine grammars integrating intelligence johnson language linguistics mark massachusetts maxent meeting methods michael michiel pages parsers parsing paul press proc proceedings references reranking richard riley roark shepherd software speech sproat stochastic treebank workshop
http://acl.ldc.upenn.edu/P/P06/P06-1007.pdf	6	A Finite-State Model of Human Sentence Processing	access account acquisition advances algorithm almost ambiguity ambiguous amherst analysis animacy antecedents approach architectures aspects asymptotically attachment bangalore basic basis bayesian behavior bever bounds brants british burnard cambridge categories category charles clause clauses clifton codes cognitive coling comprehending comprehension computational computing confer consortium constraintbased construal convolution corley corpus correcting coverage crocker decoding differences disambiguation dissertation down during earley empty ence error errors evensen evidence experimental exploring ferreira fixation formal foundations frazier frequency from geneva grammatical guide hale henderson homes human hypothesis ieee independence individual influence information inquiry international issues joshi journal juliano jurafsky just king kuczaj language learning lexical lexicalist linguistic linguistics lingusitics macwhinney making manning massachusetts mcelree mechanisms memory merlo model modeling modular morris movements naacl narayanan national natural neural object optimal oxford paced pages paola parse parser parsing parts patterns pickering predicts preference press probabilistic proceedings processes processing psycholinguistic psychology rayner reading recognition references regan relative research resolution roark role schriefers schutze science seely self semantics sentence sentences service speech srinivas statistical stevenson strategies structurally subject super supertagging supertags suzanne syntactic syntax systems tanenhaus their theory times toronto transactions traxler trueswell university users verb verbal viterbi vonk wide word working
http://acl.ldc.upenn.edu/P/P06/P06-2080.pdf	226	Semantic parsing with Structured SVM Ensemble Classification Models	allen bagging benjaming boosting breiman comparison computational conll constructing cumming decision dietterich edition ensembles experimental integrates johnson language learning linguistic linguistics machine mento methods models mooney natural park parser pcfg predictors proceedings randomization references representation semanics semantic statistical syntax that three tree trees understanding
http://acl.ldc.upenn.edu/P/P06/P06-2108.pdf	254	Using Word Support Model to Improve Chinese Input System	able aboutmsr academia accuracies achieve acoustics adaptation agent ambiguity analysis appendix application applications applying approach asian audio auto autogeneration automatically bakeoff based becker beijing bestfirst bigram boundaries broadcast case cases center chang character characteristics characters chen cheng chern chiang chien chinese chung classification coding coling colips collected commun communications computational computer computing conclusions conf conference confirmation connection constraint content conversion corpus daily data decoding denver department dictation digit directions disambiguation distance dynamic easily eijing encouraged engineers existing experimental extensible faced fact fong fourth framework fully fundations future generated given going golden goodman grammar group guan handling handwriting homophone hong hsieh http huang icassp iccpol identified identifier identifying ieee ijcnlp illustration improve incorporate information input institute integrated integrating intelligent international intl into japanese jeju joint journal knowledge koera kong korean labs language languages large level lexical liang line linguistics long machine mandarin manning markov matching method methods microsoft model modeling msime natural news nvef optimization oriental orville output pair paper parsing pattern perturbation phil phoneme phonemic phonetic pinyin poly post present press problem proc proceedings processing programming qiao real recognition references relaxation removing report research respectively results retrieval rocling satisfaction schuetze science second segmentation semantic sentence sentences sequence sheng shengzhi shiamen show sighan signal sinica speech spoken sproat srilm standard state statistical stochastic stolcke stroke study sung support survey syllabic syllable syllables symbols symposium syntactic system systems table taipei taiwan tasi task technical technique technology text that thesis this thomas through time tocharacter tonal toneless toolkit touch toward toword transaction transactions tsai tseng typing udnnews unification unified united university used using variations very vocabulary wang with word words workshop xian xing yang
http://acl.ldc.upenn.edu/P/P06/P06-2060.pdf	206	Minority Vote: At-Least-N Voting Improves Recall for Extracting Relations	adwait american analysis anlp annual applications applying approach association august bagging barcelona base based bikel borthwick breiman brill chapter chunking classifier classifiers coling combination combining computational conference connl consensus coreference cybernet daelemans damerau data david decision dejean detection disambiguation diversity driven emnlp entity entropy evaluation exploiting extraction features florian generalization halteren handwriting hassan henderson high hull human identification ieee improved improving index information intelligence ittycheriah january jing johnson journal july kambhatla koeling krymolowsky krzyzak language learning lexical linguistics machine magerman maximum meeting methods miller model modeling models multilingual multiple naacl named namefinder nanda natural nicolov nist north noun nymble page pages parse parsers parsing pattern performance phrase phrases plan predictors proceedings processing punyakanok radu ratnaparkhi recognition references research resolution roth roukos sang schwartz semantic sense soon spain speech srihari statistical suen syntactic system systems tagging technology tests text their thesis through tjong tracking trans transactions university volume weischedel winnow with word wordclass yarowsky york zavrel zhang
http://acl.ldc.upenn.edu/P/P06/P06-1133.pdf	132	Are These Documents Written from Different Perspectives? A Test of Different Perspectives Based On Statistical Distribution Divergence	abelson alexander american analysis annals answering associates association assumption automated automatic based bayes behavioral beineke belief bell bing bootstrapping brett bruce capture capturing carbonell carroll categorization chapter classification cognitive collier company computational computer computing conference conll cover criticism customer cuts data dave david detection discovery diverse document douglas ecml education elements ellen emnlp empirical erlbaum european expressions extraction fabrizio factor facts favorability forty freeman from fukushima gallery genre geoffrey goals hastie hatzivassiloglou hauptmann hinrich hong human identifying ideological improving independence individual inference information inquiry international interscience into jaime janyce kessler knowledge kullback kushal language lawrence lawrene learning leibler levels lewis lillian linguistics littman machine machines march martin mathematical matthew measuring melanie methods michael minimum mining minqing models morinaga mullen naive nasukawa natural nigel nouns nunberg opinion opinions orientation oxford pages pang pattern patterns peanut pennock perspectives peter philip plans polarity politics praise press proceedings processing product provided questions reasoning rebecca references reputations retrieval review reviews riloff ripley robert roger schank schutze science scientist scripts sebastiani semantic sentence sentences sentiment sentimental separating shivakumar side sigkdd simulation sources statistics steve stochastic structure structures stylistics subjective subjectivity sufficiency summarization summarizing support surveys systems tateishi techniques tenth text theory theresa thomas thought thumbs tois tony towards transactions trevor turney understanding university using vaithyanathan vasileios vector verdonk which wide wiebe wiley wilson with world yamanishi
http://acl.ldc.upenn.edu/P/P06/P06-4012.pdf	298	LexNet: A Graphical Environment for Graph-Based NLP	accessible according acknowledgments administered algorithmic allow allows along also andrew artificial assignable assigned attachment availability available based becomes between blue both cald centrality christopher circular clair class classification color colors corresponding data demo demos dependency designate distributions dragomir drawn each either encoded entered erkan exploiting feature fields figure file files fmeasure following foundation from ghahramani grants graph graphbased gunes http icml inducing inference information intelligence interscience jaakkola jair joachims journal kristina label labeled large larger layout learning lexical lexrank link linkbased locally machines manning markov martin membership methods metrics michigan models national nips node nodes object partially polarity precision pres press probabilistic probability problem propagation proportional radev random range read recall references remotely report reports repositories represented research rest salience schema science selected semi sentence sentences shown similarity size smaller square statistical such summarization summary supervised support supported szummer tangra technical text textual their theory this thorsten though through tommi toutanova towards transductive tumbl umich under university unlabeled user uses using vapnik vector vertex vertices very vladimir volume walk walks weakly weight while wiley with word work xiaojin york zoubin
http://acl.ldc.upenn.edu/P/P06/P06-1095.pdf	94	Machine Learning of Temporal Relations	action alex algorithms allen american analysis analyzing anders ando andrew animated approach artificial asher assocation berglund boguraev branimir chapter chklovski clare cohen commonsense compliant computational csli discourse dordrecht dublin dynamics eacl emnlp extract faller fine finegrained formalizing from general generate grover hans hitzeman http hwang ijcai inferring information intelligence internal interpretation ireland james janet johansson journal kamp kaufmann kehler kluwer kubota lapata lascarides learning linguistics logic machine marc meaning mining mirella moens nicholas north nugues ocean order pantel part patrick pauly pierre proceedings publications reasoning references relations research resolving richard ryle scenes schapire schubert semantic semantics sentence singer stanford structure swedish temporal tense text texts theory things time timeml timothy towards trees using verb verbocean
http://acl.ldc.upenn.edu/P/P06/P06-2125.pdf	271	An HMM-Based Approach to Automatic Phrasing for Mandarin Textto-Speech Synthesis	acoust algorithm analysis assigning audio automatic automatically base based beijing black boundaries breaks chen chinese chou classification communication computer contour control conversion data detect detection disyllabic durations elements english focused from fundamentals generation hall high hirschberg hwang icassp icslip icslp ieee information input instead into intonation intonational isclip iwpt kasuya klatt language level lexical linguistic mandarin mining modeling mori multi part patterns peng phrase phrasing prentice press proc processing prosodic prosody qian quality quantified rabiner recognition references review rules segmental segmenting sequences spanish speech strategy structure syllables synthesis synthesizer system taylor template text training trans tseng tsinghua uniform unit units univ unrestricted using variation vicinity wang wightman word words yang ying zhao
http://acl.ldc.upenn.edu/P/P06/P06-2031.pdf	177	Robust Word Sense Translation by EM Learning of Frame Semantics	aaai across algorithm alignment anne approach august automatic available baker barcelona base based benfeng berkeley between biframenet bilingual bilinguals boas bonnie brain burchardt cambridge canada carlos carpuat categorization chang charles chen cheung chinese classes clsp clustering coling collin common computational computing concepts conference congress construction convergent corpora corpus cortical cross daniel data dekang detour dictionaries disambiguation discrimination dong dorr editors embedded english estimating evaluation features fillmore first frame framenet fran francis frank franz french from fung geneva germanet gildea gina grace green grosjean groups hang hans hownet hsiang http huang hyungsuk iaai identifying ikeda illes indicates inducing induction integration international involving issue japanese jinying john judy july june jurafsky keenage kevin knight koehn labeling language languages ldoce lesley levow lexicon like lingual linguistic linguistics linguists lowe machine manual maps marine martha maryland matching milroy miriam mixed model monolingual montreal motor muysken ngai noun online ontological ontology palmas palmer pascale perceptual petriuck philip philipp phrase pieter ploux press probabilities proceedings processing processingvol project qiang query rebecca references representation resnik resource resources response rich roles sabine satoko semantic semantics sense shun similarity sinica skills spain spanish speaker special step stroop subitrats sujian sumo suprirse switzerland taipei taiwan task third towards translate translation university unrelated using verb wendy with word wordnet words workshop xiaohu yunbo zhendong zhibiao zhiwang
http://acl.ldc.upenn.edu/P/P06/P06-1141.pdf	140	An Effective Two-Stage Model for Exploiting Non-Local Dependencies in Named Entity Recognition	aaai abbeel accurate acknowledgments across adapt alexander also analysis applications approach approaches approximate basic bayesian belief borthwick bunescu chieu clark code coling collective complexity computationally conclusion conditional conll connections corpus curran data dependencies diego differences discriminative discussions distant distributions documents doing eacl easy efficient entities entity entropy existing exploit extraction fields finkel francisco freeman freitag gazetteers geman generalized gibbs global graphical grenager grover hidden hmms icml ieee images implement incorporating independent inference information intelligence intelligent into intractable jenny just kauffmann kaufmann koller labeling lafferty language learning leek local machine made magnitude make makes malouf manning markov master maximum mccallum mikheev model modeling models moens mooney more morgan most naacl named networks nips nonlocal often order other outperformed over pages parsing pattern pearl pereira plausible presented probabilistic proceedings propagation random reasoning recognition references relational relaxation restoration result resultant sampling segmentation segmenting sequence sequential shallow shrinkage significance simple simplicity since stage statistical stochastic sutton systems tagger taskar test tests thanks that them thesis this time transitions trond twice understand university using various very weiss well whereas while wish with without workshop would yedidia york
http://acl.ldc.upenn.edu/P/P06/P06-1042.pdf	41	Error mining in parsing results	analyse approach automatique beno berger boullier chelle computational della efficient entropy grande language langues linguistics maximun natural pierre pietra processing profonde references sagot syntaxique traitement
http://acl.ldc.upenn.edu/P/P06/P06-1076.pdf	75	A Comparison of Document, Sentence, and Term Event Spaces	adamic anking available from groups html http istl lada laws papers parc pareto power ranking references tutorial xerox zipf
http://acl.ldc.upenn.edu/P/P06/P06-3004.pdf	275	Annotation Schemes and their Influence on Parsing Results	accurate added akten alberto also amit analysis analyzing anlp anmerkungen anna annotating annotation apples appropriate approximated approximation argument bank based becomes begriff being beneficial biew both brackets brants brigitte britta broader broken charniak choice christopher clear closer collins compare comparison conclude conclusions conditioned conduct considers context contexts contrary corazza corpus covers crossing daniel data deep deficiency deleted dependencies design deutschen diesterweg drach driven drops dubey emnlp entropyinspired erhard erich especially eugene expansion experiments felder flat frank frankfurt free freguson functions further future german germanistenkongresses germany gildea giorgio gold good gottingen grace grammatical grand grundgedanken hand hans happens have head height heike helmut hinrichs hohle horizontal however human implementation independent influence influences interested internationalen italian just karen katz keller klein krenn kubler language languages lavelli less linguistic lopar macintyre main making manning many marcinkiewicz marcus mark markovization marry maximum measure michael mitchell mittelfeld models more moreover naacl natural negra node number observed oranges order outlook parent parse parser parsers parsing particularities pcfg pcfgs penn pennsylvania performance plainsboro point predicate presented probabilistic probabilities probability problem proceedings provide ranlp reduce references reliable report results right robert roberto rule rules sandra satta satzlehre schasberger scheme schemes schmid seen seminar serious show sides siebten since sisterhead skut sparseness sprachwissenschaft stanford start state statistical step structure structures stuttgart stylebook subtree tags technical technology telljohann than that theorie theories thesis this thorsten tilman time topologischen tree treebank treebanks tuba tubingen universitat university unlexicalized uses using uszkoreit variation verb vertical when which will wise with wojciech word work workshop written zanoli
http://acl.ldc.upenn.edu/P/P06/P06-2066.pdf	212	Mildly Non-Projective Dependency Structures	abeill ability academic account advances albany alena alexis algorithm algorithms anne annotation annual anssi approximate association barbora based bodirsky brill building bulgarian capacity chapter charles christoph coling collins computational computer conference constraint constraints contextfree control corpora corrective czech daniel danish datadriven davy debusmann delaware denys dependency dick discrete drawings dtag duchier eacl editor efficient eigth eisner electronically eleventh encyclopedia english eric eryigit european exploration fernando formal fourth gaifman generative giorgio gralinski grammar grammars haim hajic hajicov hall hladk hladka hmov holan home http hudson igor information integer international intro iwpt january jarmila jason jens joakim kahane keith kemal kiril kluwer kromann kubon kuhlmann lance language learning level line linear linguistic linguistics link lsen manuel marco marinov martin mathematics mathias matthias mcdonald meeting michael model modeling models multiplanarity nasr neil nested newark nilsson ninth nivre njas notes obrebski oflazer online order owen pages pajas panevov parsable parsed parser parsing pereira petr phon phrase polynomially practice prague precedence press probabilistic projective projectivity pseudo pseudoprojective published publishers ralph rambow ramshaw recent recognition references relax republic research rewriting ribarov richard ryan satta scenario science second sequences sgall sleator sloane some spanning state statistical structure structures svetoslav sylvain syntactic syntax systems technologies temperley tenth theoretical theories theory thesis third three tillmann tool topological trautner tree treebank treebanks trees turkish university using vaclav vidova vladislav well with word workshop york zeman
http://acl.ldc.upenn.edu/P/P06/P06-1136.pdf	135	Reranking Answers for Definitional QA Using Language Modeling	altavista annotation annual answer answering applied approach association automatic banko based biterm blair bleu brazil brill ceedings center chen chiba chin chua coling computational conference croft czuba data deepak definitional definitions dependence development dictionary diego document dumais eduard effectiveness eleventh ellen empirical evaluation experience exploiting fduqa finland gaithersburg general goldensohn goodman guangyuan guihong hang harper hazen hidden hovy http huang human hybrid industrial information integrating intensive international into japan jian jianfeng jing jinxi kraft lafferty language learning leek licuanan lide linguistics louisiana machine magnini markov mckeown meeting method methods miller minimization model modeling models negri nist orleans overview papineni passage patterns philadelphia ponte practical prager prevete proceedings query question questions radev ralph ranking ravichandran redundancy references relationships report reranking research retrieval right risk roukos salvador schlaikjer schwartz seng sheffield sigir smoothing soft song srihari srikanth study supervised surface system taipei taiwan tampere tanev techniques technology tenth text thirteenth thomas track translation trec twelfth unsupervised validation virtual voorhees ward watson weischedel what wide with word world xuanjing yaqian york yunbo zhai zhang zhao zhou zhushuo zobel
http://acl.ldc.upenn.edu/P/P06/P06-1013.pdf	12	Ensemble Methods for Unsupervised WSD	about acapulco accuracy accurate across alfio algorithms annotation antal approach arpa automatic banerjee barcelona based bases biewald bosch briscoe bunker canaria carlo carroll chaining charles christopher class classifiers claudia claudio cohesion combination combining computational computed concordance cone cream cucerzan current daelemans daphne data david definition dekang designing determining diana dictionaries dietterich differentiating directions disambiguation diverse dominance eacl editors edmonds emnlp engineering evaluating extended extension finding flairs florian florida four from galley general george getting giuliano gliozzo gloss graeme gran graph halteren hans hendrickx hirst homonymy hoste hwee icml ijcai improving indicator information interconnections iris italy jakub jane john julie kathleen kaufman knowledge koeling koller labeling language large largescale leacock learning lesk lexical linguistic linguistics lrec luke machine madison magazine marc mccarthy mckeown measure michael michel mihalcea miller mohammad morgan morris multiple natural navigli ning note opinions optimization ordering overlaps pages palmas pami pang paola parameter patterns pedersen phil philip pine polysemy predominant proceedings rada radu randee readable references relatedness relations research retrieval roberto robust rong ross saif satanjeev schafer seattle semantic semantics semi sense senses senseval sequence serious sigdoc siglex silviu similarity sources spaces spain statistical stokoe strapparava structural structure systems tagging task technical tell tengi text teyssier theoretic thesaural thesaurus through translation trento unsupervised untagged using vancouver velardi veronique vickrey vocabulary walter washington weeds what with word workshop yarowsky york zavrel
http://acl.ldc.upenn.edu/P/P06/P06-1040.pdf	39	Expressing Implicit Semantic Relations without Supervision	acquisition agichtein analysis annual association automatic baltimore berland brin charles charniak christopher clarke coling collections communications computational computations conference cormack corpora database digital disambiguation dumais edbt edition english eugene extending extracting fifth finding forum from gene george golub gordon gravano hearst heidi hopkins hyponyms induction international johns knowledge lance landauer lapata large latent lexical libraries linguistics loan luis maria marti matrix matthew meeting miller multitext nominalisations novel overview pages palmer pars parts patterns plain plato press problem proceedings psychological ralph ramshaw references relations representation review scott semantic sergey sigir snowball solution statistical susan technology text theory third thomas university very webdb weischedel wide wordnet workshop world
http://acl.ldc.upenn.edu/P/P06/P06-1126.pdf	125	Discriminative Pruning of Language Models for Chinese Word Segmentation	alexander andi annual approach association backoff better chang channel chinese clarkson cmucambridge collins communication computational conference criteria discriminative distributionbased entity european eurospeech fredrick huang icml improved improving international jelinek jianfeng language learning linguistics machine meeting michael model modeling models named natural ning organized pages parsing philip pragmatic proc pruning recognition reduction references reranking ronald rosenfeld segmentation self size source speech statistical technology toolkit using waibel word zhang
http://acl.ldc.upenn.edu/P/P06/P06-1134.pdf	133	Word Sense and Subjectivity	aaai about adjectives affective algo analysis analyzer annotating answering applied approach attitude automatic available based budapest bunescu canada cardie carroll categorization chklovski cikm classification clustered clustering coling collecting communication comprehensive computational conference conrath corpora corpus crosscultural customer cuts database dave determining disambiguate disambiguation down education electronic emnlp emotions empirical english esuli etzioni evaluation examplar exploiting expressions extracting extraction facts features finding from gallery given gloss grammar greenbaum hatzivassiloglou heise hovy http hungary icdm identifying improve information integrating international internet jiang journal kamps kilgarriff knowledge koeling language lawrence learning leech lewis lexical linguistics longman magellan marx mccarthy mckeown meanings mihalcea miller minimum mining montreal multiperspective multiple nasukawa natu niblack onomy opinion opinions opqa orientation pages pang patterns peanut pennock phillips phrasal polarity popescu predicting predominant proc proceedings processing product project question questions quirk references representations research resources retrieval reviews riloff rithms sample sebastiani semantic sense senses senseval sentences sentiment sentimental sentiments separating sigir sigkdd siglex similar similarity sociology sources statistics stoyanov subjective subjectivity summariza summarizing svartvik taiwan task techniques terms text through thumbs tion topic towards traction turney unsupervised untagged using weeds wiebe wilson with word wordnet words york
http://acl.ldc.upenn.edu/P/P06/P06-4016.pdf	302	TwicPen : Hand-held Scanner and Translation Software for non-Native Readers	accessing advances amsterdam analysis antecedent assistant based been benjamins bigrams break breidt broke chain coindexed coling collocation compass composition conclusion containing context csli demand developed development dialog direct display dokter easy either empty extraction face feldweg foreign fragment from gawron given global glosserrug identify increase intelligent john language languages learning lecture likely line machine marlien material mots multi multilingual natural nerbonne nerima nicolas nicolov norvig notes notice noun object orleans paper papers phrases presented printed proceedings processing pronoun ranlp readers reading recent record recordi references relative relatively scan selected sentence seretan smit societies stanford summit support syntactic system taln terminological that thati they this toface tools trace traduction translation turn twicpen understand verb verbmobil wehrli which will with word words
http://acl.ldc.upenn.edu/P/P06/P06-4014.pdf	300	Re-Usable Tools for Precision Machine Translation	already available bond carroll chart copestake delph delphin details efficient european flickinger france generation generator grammars http including language lexicalist lists machine natural oepen open participating phuket poznanski proceedings references resources semi siegel sites source summit thailand toulouse translation with workshop
http://acl.ldc.upenn.edu/P/P06/P06-2098.pdf	244	Exact Decoding for Jointly Labeling and Chunking Sequences	advances aggressive algorithm algorithms biology classification collins computational crammer dekel detecting diekhans discriminative emnlp empirical experiments framework freund haussler hidden homologies information jaakkola journal language large learning machine margin markov mcdonald methods models multiclass natural neural nips online outputs passive perceptron pereira prediction problems proc processing protein references remote research schapire shalev shwartz singer structured systems theory training ultraconservative using with workshop
http://acl.ldc.upenn.edu/P/P06/P06-2010.pdf	156	A Hybrid Convolution Tree Kernel for Semantic Role Labeling	accurate algorithm approach argument aron automatic baker berkeley cambirdge cambridge carreras charles charniak chieu classification coling collin collins computational conll convolution cristianini culotta daniel david dependency discrete duffy entity entropy entropyinspired eugene extraction fast fillmore framenet freund gildea gimenez haussler hwee introduci introduction jeffrey jesus john july jurafsky kernels labeling language large learning leong linguistics lowe machines margin marquez martha maximum michael naacl named natural necessity nello nigel nips pages palmer parser parsing part perceptron predicate press proceedings project ranlp recognition references relation report revisited robert role roles schapire semantic shared shawe sorensen speech structures support tagging task taylor technical theory tion tree ucsc university using vector with xavier yoav
http://acl.ldc.upenn.edu/P/P06/P06-1112.pdf	111	Exploring Correlation of Dependency Relation Paths for Answer Extraction	acoustics adam adrian albany algorithm algorithms alignment answer answering answers approach approximate based becker berger broadcoverage calculated candidate chains chua classifier coling combining computational conclusion considerations corpora correlation correlations deepak dekang della dietrich difference discrete duan dynamic each eduard efficient entropy exploring extract extraction franz from geert hang hovy ieee ijcnlp ilqua incorporate irst josef kaisser keya klakow kouylekov kruijff language large lastly levinson lexical linguisitic linguisitics linguistic magnini mapping maximum measure method methods mining model moldovan multilingual national natural next nist novischi pages pair paper parser path paths patterns phrase pietra principar principle proceedings processing propose proposed question questions rabiner rank ranker ranking ravichandran recognition references relation renxu rosenberg score searching seng sentences sequence shaikh shen signal singapore small speech statistical stephen strzalkowski summarization syntactic tanev this time transactions trec university vincent warping what which with word workshop
http://acl.ldc.upenn.edu/P/P06/P06-1028.pdf	27	Training Conditional Random Fields with Multivariate Evaluation Measures	altun approach backpropagation basis carnegie categorization chua cost cybernetics determination discriminative emnlp empirical expected fahlman figure formal functions hart heuristic hofmann ieee investigating jansche johnson label learning logistic loss maximal maximum measure mellon merit methods minimum models networks nilsson optimization pages paths proc raphael references regression report science sequences sigir speech study systems technical text training trans university
http://acl.ldc.upenn.edu/P/P06/P06-4007.pdf	293	F E R R E T: Interactive Question-Answering for Real-World Environments	acquisition addition analysis andrew annual answering answers association automated automatically bipartite bowden chain chin chua clark coling combines complex computational conference contributions decomposed decomposition decompositions development displayed each eduard employing experiments finley following fourteenth from geneva germany graph harabagiu hickl hovy however improves incremental information interactive international jiang john lacatusu lehmann linguistics markov meeting models moldovan notified occurred operates performance presents procedure proceedings processing question questions random references relation relations relevant representations represents request research retrieval saarbrucken sanda seattle semantic semantically separate sigir signatures single strategies summarization switzerland syntactic system systems table techniques text that topic trec types unlike upon user users using variety walk wang with
http://acl.ldc.upenn.edu/P/P06/P06-2100.pdf	246	Morphological Richness Offsets Resource Demand- Experiences in Constructing a POS Tagger for Hindi	acai accuracy agglutinative agrawal aided algorithm analysis annotation annual applied arabic automatic based basu bharati bhattacharya black brill bryant case chaitanya chapter christodoulakis clark claws coling combined comparing computational computer conf conference corpora corpus csirik cutting czech darpa decision dictionary disambiguation distribution driven emnlp entropy error european figure foundations from garside grammatical grouping guiassa hajic hall harish harnessing hcza hindi hungarian hybrid icon ifie improving india induction information isahara japanese joint journal kalles krbec kuba kuruoz kveton labeling language learning leech linguistic linguistics local longman machine manning manual maximum mcenery meeting megyesi method methods models morphological natural networks neural niblett oflazer oliva orphanos osit pages paninian papagelis parsing part partof parts perspective petkevic postagging practical prentice press problem proc proceedings processing procs ratnaparakhi references rulebased samuelsson sangal sarkar schmid schutze science sekine shrivastava sigdat singh smith speech statistical stochastic study tagger tagging task techniques text third tlili transformation tree trees turkish uchimoto unknown using voutilainen with word workshop
http://acl.ldc.upenn.edu/P/P06/P06-2124.pdf	270	BiTAM: Bilingual Topic AdMixture Models for Word Alignment	algorithm allocation americas amta andrew application approach approximation association based bayesian beal beam blei bonnie brown carpua cohesion computational conference data david decoder dekai della dirichlet disambiguation dorr empirical estimation evaluating fifth generation ghahramani graphical habash heavy heidi ijcnlp incomplete interlingua international joint jordan journal july koehn language latent learning linguistics machine marine mathematics mercer methods model natural nizar pages parameter performance pharaoh philadelphia philipp phrasal phrase pietra proc proceedings processing references reliability research robert scoring search second sense statistical statistics stephen structures tiburon translation variational vincent volume with word workshop zoubin
http://acl.ldc.upenn.edu/P/P06/P06-2109.pdf	255	Trimming CFG Parse Trees for Sentence Compression Using Machine Learning Approaches	aaai accuracy allows approach appropriately automated automatic based because berger bigram bleu bottom bottomup branches cannot characteristics charniak coarse complicated compressed compression computational conclusion condensed corpus correctly could criteria decomposition della depth described determined determining discriminative dorr eacl entropy evaluation evidence experimental extend extractive fine first fold forest from further generating generation grammatical headline hedge highly however human hybrid iaai improved incorporate informative input into jing johnson judgments knight langkilde language learning linguistics machine marcu match matched matching maxent maximum mcdonald mckeown measures method methods mittal model modify more mother naacl natural nbest need node original other package pages papineni parse parsing part pietra presented previous probabilistic proc process processing produced proposals provided references removed reranking results root rouge roukos schwartz scores second sentence sentences short show showed sigir small soft some statistical statistics step structures subtitling such summaries summarization summarized summary supervised syntactic text than that this though three training translation tree trees trim trimmer turner ultrasummarization unsupervised using vandeghinste various ward were which witbrock with workshop written zajic
http://acl.ldc.upenn.edu/P/P06/P06-2081.pdf	227	Whose thumb is it anyway? Classifying author personality from weblog text	accessed acquaintance adrian affect alastair america annual applied arat argamon assessment austin author authorship automatically banerjee buchanan buchant cambridge categorizing city classification computational computing conference costa dave david deary design dewaele dhawle differences discrimination edition electronic elizabeth extraction extraversion factor first five fourth furnham gallery gender gerald gill henry html http hugo implementation individual intelligent interface interfaces international introduction inventory ipip james jean joint knowledge koppel kushal language lawrence learning lexical lieberman linguistic linguistics literary mail manual marc marin martha matthews mccrae meeting messages mexico mining model moshe multiple ngram north oberlander odessa online opinion package page pages paul peanut pedersen pennebaker pennock personality predictors press proceedings processing product professional psychological rating real references research resources results reviews revised robert saric satanjeev selker semantic sensing shimoni shlomo sigkdd society statistics stein sterling steve style sushant text texts textual traits type university unloved user users using variable whiteman wide wmin world written wwwffi zero
http://acl.ldc.upenn.edu/P/P06/P06-1009.pdf	8	Discriminative Word Alignment with Conditional Random Fields	achieves acknowledgements affect alberta algorithms aligned aligner alignment alignments allow allowed amount anonymous approach approaches arabic arbitrary arbor assess association audio baldwin barcelona based baseline between beyond bilingual bird both british brown building burch callison canada case characteristics chen columbia comments comparison computational conclusion conditional conf conll corpora corpus could data della demonstrated derived dice discriminative driven easily ecologic ecology edmonton efficient emnlp employ english entropy error estimation evaluated evaluation exercise expected extentions extracted features feedback fields framework french from further furthermore generative have heuristic highest highly hundred icml ieee ilhan improve improving incorporated inducing inference insightful into ittycheriah journal julien july june klein koehn labelling lacoste lafferty languages linear linguistics machine making malouf manning many marcu martin matching mathematics maximum mccallum measures mercer methods michigan mihalcea miles model models moore moreover most naacl necessarily novel number october only optimality osborne outperforms over overlapping pages paper parallel parameter parrallel parsing pedersen pereira philadelphia phrase phrasebased phrases pietra plan precision predictive presented previous probabilistic proceedings processing quality random rate recall references reported resources results reviewers romanian rosenfeld roukos sacrificing scarce segmenting sentence sentences sequence sets shallow showed small smoothing source spain special species speech statistical steven still supervised survey systematic table talbot target taskar techniques template terms texts thanks that their therefore these this tillmann timothy tolga toutanova tradeoff training transactions translation used using vancouver various vogel were when while with without word wordaligned work workshop
http://acl.ldc.upenn.edu/P/P06/P06-1084.pdf	83	An Unsup ervised Morpheme-Based HMM for Hebrew Morphological Disambiguation	allon approaches arabic architecture associated automatic base based baum beer brill building cambridge carmel case chen choosing chunks computational daniel david diab disambiguation emmanuel entities eric errordriven estimation from functions gurion hacioglu haim harvard hebrew inequalities inequality israel jurafsky kadri khalil language languages languge learning leonard linguistics maarek markov master maximization models modern mona mordechai morphological naacl named natural negev ngits optimal part phrase press probabilistic proceeding proceedings process processing recognition references search segmentation semitic sheva sima speech stanley statistical study systems tagging technique text thesis transformation university unvocalized winter workshop writing yoad yoelle
http://acl.ldc.upenn.edu/P/P06/P06-3013.pdf	284	Extraction of Tree Adjoining Grammars from a Treebank for Korean	adjunction adverbial also arguments between candito consider contain covered difference extracted frames frequency generally grammaire hakgyo information labeled lectronique lexicalis lexicalized like marie marked modulaire morning neun noun number ojeon only operation organisation param paris phenomena phrase phrases prepositional presents references schemata school since sjtree some subcategorization syntactic there thesis this three trable tree trees universit variations went which while with
http://acl.ldc.upenn.edu/P/P06/P06-1107.pdf	106	Discovering asymmetric entailment relations between verbs using selectional preferences	advances annual association barcellona based borovets bulgaria canada case chklovski church computational conference corpus dagan empirical entailment fine from glickman grained hanks identifying information international kenneth koppel language lexical lexicography linguistics meeting methods mining moshe mutual natural norms oren pantel paraphrases patrick probabilistic proceedings processing ranlp recent references relations semantic single spain study textual timoty vancouver verb verbocean verbs ward word
http://acl.ldc.upenn.edu/P/P06/P06-1117.pdf	116	Semantic Role Labeling via FrameNet, VerbNet and PropBank	advances alessandro annual artificial automatic baker based berkeley burges carreras case charles charniak classes coling collin computational conference conll dang daniel discovering ecml editors entropyinspired eugene european extensions fillmore framenet framenets frames garda gildea giuglea intelligence intersective introduci investigating italy joachims josef joseph jurafsky karin kernel kipper knowledge labeling large learning levin levins linguistic linguistics making maria marquez martha maximum meeting methods moschitti nacl ontology palmer parser parsing pisa practical proceedings propbank references regular riva role roles rosenzweig ruppenhofer scale scholkopf seattle semantic sense shallow shared smola society support task theory tion trang universals using vector verb verbnet washington workshop xavier
http://acl.ldc.upenn.edu/P/P06/P06-3003.pdf	274	Sub-sentential Alignment Using Substring Co-Occurrence Counts	accessing advantage alex algorithm aligned alignment alignments allowed also although applied arbor array association automated bannard barcelona base based beijing better both brown btec burch callison characters china chinese chris christophe colin commonly computational computing conclusion conference considered corpora corpus count counting create daniel dictionary draft empirical engineering equivalence europarl evaluation example extraction franz free french future hardly hermann however ideas improved infrequent integrated interesting international issues joint joseph josh journal july knowledge koehn language large larger learning like line linguistics longer machine madrid make manber marcu maryland melamed method methodological methods model models mostly multilingual myers natural need occurrence occurrences october pair paper parallel performs philadelphia philipp phrase phrases presented probability proceedings processing produced rais ralph references research results richard santa satisfying scaling scroeder searches seemingly segmentation showed siam sighan some spain statistical stephan string subsentential subset substring such suffix taking than theoretical think this thorough tillmann tool translation translational unit university unpublished used useful variations very vogel voud voudrais waibel well were where william with wong word work workshop would ying zens zhang
http://acl.ldc.upenn.edu/P/P06/P06-1097.pdf	96	Semi-Supervised Training for Statistical Word Alignment	abraham alexander aligned aligner alignment annual arabic artificial association available barcelona basu bilenko brown burch california callison cherry chris clustering colin computational computers conference corpora daniel data david dekang della discovery english entropy estimation framework fraser fred future glover html http human improve integer intelligence international ittycheriah japan july knowledge language linguistics links machine marcu mathematics maximum measuring meeting mercer mikhail miles mining model mooney operations osborne pages parallel parameter paths peter pietra press probabilistic probability proc programming quality raymond references report research roukos salim sapporo semisupervised sentence sigkdd southern spain statistical stephen sugato talbot technical translation university vincent with word york
http://acl.ldc.upenn.edu/P/P06/P06-2038.pdf	184	Speeding Up Full Syntactic Parsing by Leveraging Partial Parsing Decisions	advances american annual anti arivind association based bigram boosting carreras chapter chark charniak clause collins computational conference dependencies driven ecml editors email entropyinspired eugene european filtering first fourth francisco head identification inference international john joshi kaufmann language learning lexical linguistics london machine marcus marquez martha maximum meeting michael mitchell models morgan natural north pages palmer parser parsing pennsylvania proceedings processing publishers punyakanok ranlp recent references roth spam springer statistical supervisor thesis thirty trees tzigov university vasin verlag xavier
http://acl.ldc.upenn.edu/P/P06/P06-3007.pdf	278	Investigations on Event-Based Summarization	annual applications association automatic available barzilay bontcheva brain brandow chains computational condensation conroy cunningham development diana document eacl electronic elhadad environment framework gate graphical hamish html http information intelligent john judith kalina karl left lexical linguistics lisa management maynard meeting michael mitze multi nist proceedings processing publications pubs references regina right robust ronald scalable schlesinger selection sentence summarization tablan text tools using valentin workshop
http://acl.ldc.upenn.edu/P/P06/P06-2082.pdf	228	Analysis of Selective Strategies to Build a Dependency-Analyzed Corpus	active analysis analyzer annotation annual applications applicatoins association atlas atsushi audio automatic baldridge banko base based brants brill cascaded case christopher chunking cohn coling computational conference conjunctive conll corpora cost dagan data david dependency detection dilek disambiguation efficient emnlp engelson eric example flickinger from fujii generalization gerard giuseppe grace hakkani hozumi human ieee improving inui japanese jason kamm kentaro kristina kudo kurohashi ladner language large learning linear lingo linguistics long machine makoto manabu manning manual matsumoto meeting method meyer michele miles minimizing motivation nagao natural ngai noun oepen osborne pages parser parsing phrase post preliminary proceedings processing rebecca recognition redwoods references resource resources riccardi richard rule sadao sample sampling sasano scaling sean selection selective sense sentences sharable shieber speech statistical stephan structure structures stuart supervised syntactic takenobu taku tanaka technology teresa theory thorsten time tive tokunaga total toutanova training transactions treebank usage using very with word workshop workshops writing yarowsky yuji
http://acl.ldc.upenn.edu/P/P06/P06-2069.pdf	215	Examining the Content Load of Part of Speech Blocks for Information Retrieval	academic across alan also amati annotated appears application associated association based baseline bear bears beatrice beaulieu been benefit between bhavani block blocks both brown bruce building butterworths cambridge categorization christopher church class closed clustered coling collections combination combined compare components computational concept conclusion contains content context corpus croft data david della described desouza difference differences different discovery distribution divergence division document documentation dordrecht down effect english estimators evaluated evaluation expansion extraction firstly five following foundations from further future gatford general gianni glasgow gram guided hanks have helping higher highfrequency hinrich hypotheses hypothesised identical improved improvement improvements indicate indicating indicative induced information ingrid interpretation investigate jennifer john jones journal karen keith kenneth kickoff kluwer lafferty language large lead least less lewis lexical lexicography linguistics london longer lowfrequency lyons manning marcinkiewicz marcus mary mercer micheline mike mitchell modeling models more most mutual natural nist noise norms observe observed okapi ones other outcomes over overall paraphrasing part patrick payne penn performance peter phase phrasal pietra press previously probabilistic processing provement publication publishers queries query randomness raskutti recapitulate reduced reduction references report reported representations resources results retrieval rijsbergen robert robertson robust santorini schemes schutze secondly semantics settings sharp shorter sigir significant similar size smaller smeaton smoothened sometimes sparck sparseness special specificity speech springer standard statistical statistically stephen steve strzalkowski summarization table task tasks term test testing text than that these thesis this times tipster tomek tones trec treebank university user using valid varying vincent volume walker weaker weighting well when where which while wish with without word worked workshop zukerman
http://acl.ldc.upenn.edu/P/P06/P06-1144.pdf	143	Multilingual Document Clustering: an Heuristic Approach Based on Cognate Named Entities	algorithm arantza benoit besancon casillas chen christian chuan cicling clustering clusters computational computer conference discovery document feature fluhr genetic gonzalez hsin intelligent international lecture lena linguistics mart mathieu multilingual news notes proceedings processing raquel references riao romanic sampling science selection springer summarizer teresa text verlag
http://acl.ldc.upenn.edu/P/P06/P06-1089.pdf	88	Guessing Parts-of-Speech of Unknown Words Using Global Information	above accuracies adam addressed adwait aided algorithm algorithms allows also although ambiguous analysis andrei andrieu applied approach arnaud asahara attempted automatic baayen based bayesian berger between bfgs cambridge carnegie category chao chen cheng chieu chinese christodoulakis christophe christopher cmucs coling computational computers conclusion considering consistent context contexts corpora corpus dale data david decision della dictionary dimitris directed directly disambiguation discussed distribution distributional doctor document documents dong doucet eacl easily emnlp entity entropy equation especially essentially estimated estimating estimation estimators experimental exploiting exponential expressive extracting extraction feature features fields finkel form forms framework freitas frequency from function gaussian generalized gibbs giorgos gives global greiner grenager guessing handle harald have hiroya hitoshi hong hwee icml icslp image improvement improves including incorporate incorporating induction inference information initial institute integration interactions interpolation into introduction inui isahara iterative japanese jenny jiann jordan jorge jose kiyotaka klakow language large learning leong lexical limited linear linguistic linguistics local machine mackay makoto manabu manning markov marroquin masaki masayuki mathematical maximum mcmc mellon memo memory method methods michael mikheev ming model modeling models mori morphological morphologically morphology nagao nagata named nando nara natural nlprs nocedal note obtained ofspeech okumura optimal optimization orientations orphanos other pages paper parameters parsers part parts pietra presented press prior priors problem proceedings processing programming random ratnaparkhi recognition reconstruction references regard regularities report results richard rivaling ronald room rose rosenfeld rule russel same sampling satoshi scale scaling schuurmans science scope segmentation sekine semantic semisupervised sense sentence sentences shallow shaojun shaomin shinsuke showed smoothing some speech spin sproat stanley statistical stephen such supervised surface syntactic systems tagging tags takamura takashi tasks technical technology that their theory there thesis they this training trees trigram trond uchimoto university unknown unknownword unlabeled unsupervised used using vehicle vincent wang well were where which whole wide will with word words xiaojin yarowsky
http://acl.ldc.upenn.edu/P/P06/P06-1068.pdf	67	A Study on Automatically Extracted Keywords in Text Categorization	aaai advances aizawa akiko algorithms anette automated automatic burges caropreso case categorization cikm classification combining computer conference databases david department document dumais editors ellen emnlp empirical evaluation extensive extraction fabrizio feature fernanda forman furnkranz george given heckerman hulth improve improved independent inductive information international joachims johannes john journal kernel keyword knowledge language large learner learning linguistic machine making management maria matwin mehran methods metrics mitchell more natural nlprs pacific pages performance phrases platt practical practice press proceedings processing references representations research riloff sahami scale scholkopf sciences sebastiani selection seventh smola stan statistical stockholm study support susan symposium systems techniques text theory thesis thorsten university usefulness using vector workshop
http://acl.ldc.upenn.edu/P/P06/P06-1029.pdf	28	Approximation Lasso Methods for Language Modeling	adaptation algorithm algorithms analysis angle applied approach approximation asymptopia bacchiani backward baseline berkeley bigram blasso boosted boosting both brian classification clustering collins combining comparative computational conclusion confidence consequence corrective curves dataset david department dependency discriminative discussion domain donoho dual duda efficient efron elements emnlp empirical encarta entropy error estimation evaluation experimental exploiting feature features figure fixed freund friedman graphical hart hastie have headword hisami icassp icml ijcnlp improved including input investigates iyer japanese jianfeng john johnstone journal kerkyachairan language large lasso learning least linguistics machine machines methods metric metrics michael minimum model modeling models murat naacl natural nikkei nips norm number numerical osborne outperforms over paper parsing pattern perceptron performance peter picard predictions predictive preferences presnell press problems promising rated realistic reduction regression report reranking results richard risk roark robert rosset royal sample saraclar schapire selection settings shincho shown shrinkage significantly similarity singer size sons space sparse springer squares statist statistical statistics step stork study superior support suzuki table task tech terms terry test text that this tibshirani trained tuneup turlach update using variable vector verlag very vocabulary wavelet weight wiley with yang yomiuri yoram york yuan zhao
http://acl.ldc.upenn.edu/P/P06/P06-3001.pdf	272	A Flexible Approach to Natural Language Generation for Disabled Children	acknowledgement acoustics aids annual applied approach arnott artificial asia association augmentative author available banerjee bangalore based bases becker both busemann callaway cerebral charles children communication communications composition computational computer conclusion conference conversation conversational corpus correct deemter designing dialog different disabled discusses during ernet evaluated evaluation exchanging expanding exploits expressiveness false first flexible following foundations fourth framework from generation gram grammatically grateful guide guided hovy http human ideas ieee iitkgp ilan increase indian inherent initial institute inteligence intended intention interaction interfaces interlingua international irene james june kees kevin keystrokes kharagpur knight knowledge kolkata laboratory langkilde language large lester liang linguistics machine marilyn mccoy media meeting michael minimum models momentum nathalie natural newell number october opposition owen palsy paper part parts pasero paul people performance picheny plan point pragmatics prediction present proceedings processing project projects providing quadriplegic rambow rate realize references reordering research resources respects results richardet robert robust sabatier sanyog scale sentence sentences show signal simple speech srinivas start statistical stephanidis symposium system techniques technology telegraphic templatebased that thesis this transactions translation user users views walker will with work yuqing
http://acl.ldc.upenn.edu/P/P06/P06-2119.pdf	265	Word Sense Disambiguation using lexical cohesion in the context	barzilay cambridge chaffin chains cohesion college context correction database detection electronic elhadad english fellbaum halliday hasan hirst intelligent ists lexical lexicon london longman madrid malapropisms mental onge organization paradigmatic press references representations scalable spain state summarization text trenton using verbs wordnet workshop
http://acl.ldc.upenn.edu/P/P06/P06-1135.pdf	134	Improving QA Accuracy by Question Inversion	analysis answering approach askmsr banko brill carroll clarke complex cormack czuba data developing dumais emnlp experiments ferrucci hendrix interface kisman language lynam multi multitext natural passage prager proceedings question references sacerdoti sagalowicz selection slocum source strategy system trec vldb welty
http://acl.ldc.upenn.edu/P/P06/P06-4019.pdf	305	Outilex, a Linguistic Platform for Text Processing	actes algorithm anne annotation appear architecture atala automata automates automatic automatique avec based blanc borovets brill bulgarie case cedrick clement clergerie comm companion computational computers conference constant context cunningham danlos dister driven earley ecrits editor editors efficient engineering eric error expletive fairon finite forme framework free french gate general grammars graphs gross hamish hoey humanities incoma international jeju joint korea language langues laporte laurence learning leuven lexicalization lexicaux linguistics lionel local logicielle matthieu maurice mertens morphosyntactic natural naturelles occurrences olivier outilex page pages parameterized parsing part piet plate poland poznan proc processing pronoun ranlp recital recognition references representation september speech structure study tagging taln technology text textes their traitement traitements traits transformation volume with
http://acl.ldc.upenn.edu/P/P06/P06-1131.pdf	130	Incremental generation of spatial referring expressions in situated dialog	activation angle annual april artificial assignment associates association berlin beun bielefeld bryant cambridge campbell carlson charles clark cognition cognitive collaborative composition computational conference confernce conklin construction conversation cremers csli dale david definite deixis described descriptions discourse distance domain during duwe editors edwards english erlbaum european expressions external ferrier fifth fillmore frame frameworks franklin functional gapp gardent garrod generating generation geometry gibbs gorniak gricean grounded haddock hajicova herskovits images influence intelligence interdisciplinary internal international interpretations investigating involving irwin issues jeffrey journal kommunikatoren kunstliche language lawrence lecture linguistic linguistics logan maxims mcdonald meeting memory mental minimal model moulin natural object oliver pages patterns pragmatics prague prepositions press problem proceeding proceedings process processing projective publications radvansky reference references referring reiter relations relationship report representation representing research salience scenes science selection semantic sentence shape shared simulation situierte society spatial strohner structure studies study template term their theoretical towards tversky univeristat university using visual volume vorono wilkes
http://acl.ldc.upenn.edu/P/P06/P06-2045.pdf	191	A Collaborative Framework for Collecting Thai Unknown Words from the Web	algorithm anlp applied asahara boyer categorizing chang characterbased chinese chunking coling communications computational conference decision fast identification identify international iterative janine japanese jing language lexicon linguistics masayuki matsumoto method misspellings moore names natural proceeding proceedings processing references searching shin string toole trees unknown unsupervised using word words yuji
http://acl.ldc.upenn.edu/P/P06/P06-3002.pdf	273	Unsupervised Part-of-Speech Tagging Employing Efficient Graph Clustering	accurate against algorithm amsterdam application approach attempt based biemann bootstrapping categories categorisation charniak chater chinese claim clark cluster clustering coincidence coling combining computational computer conclusion conference corpus decision dimensionality distributional dongen dunning eacl efficient equations feature finch finland first freitag further geneva granularity graph graphs hendrickson independence induction information institute international jacobson joensuu knowledge language languages leave linguistically linguistics manchester mathematics menlo methods morphological motivated naacl national natural netherlands nodalida output pages park part perkowitz presented probabilistic problems proc proceedings processing reduction references report research rigorous schmid science selection shoe speech statistical statistics supervised supported surprise syntactic system systems tagger tagging technical text textgraphs this three through tilburg toward trees unsupervised using validating whispers whole witschel work workshop york
http://acl.ldc.upenn.edu/P/P06/P06-1142.pdf	141	Learning Transliteration Lexicons from the Web	algorithm anatomy approach automatically brill brin brockett coling companion comparable computational constructing corpora data dempster emnlp engine english from fung graehl hall harvesting huang hypertextual incomplete jersey journal jurafsky kacmarcik katakana knight laird language largescale lexicons likelihood linguistics logs machine martin maximum mining nlpprs nonparallel page pairs phrase prentice proc processing query references royal rubin search society speech statistical stephan term texts translating translations transliteration transliterations vogel volume words yang zhang
http://acl.ldc.upenn.edu/P/P06/P06-2022.pdf	168	Automatically Extracting Nominal Mentions of Events with a Bootstrapped Probabilistic Classifier	acquisition anlp aone approximate artificial baker berkeley briscoe celex centre coling conf copestake corpus curran dahl database deep distributional edition english event extension extraction ferro fillmore framenet gaizauskas gomez gorman hanks hull information intelligence interpretation journal lancaster large lazo lexical linguistics lowe national nijmegen nominalizations online pages palmer passonneau polysemy proc productive project pundit pustejovsky ramos rees references relation santacruz sauri scale searching semantic semantics semi sense setzer siglex similarity sundheim system timebank workshop
http://acl.ldc.upenn.edu/P/P06/P06-1086.pdf	85	M AG E A D : A Morphological Analyzer and Generator for the Arabic Dialects	american analyser analysis analyzer approaches arabic arbor association beesley bilingual bird buckwalter building charles comprehensive computational computing darwish dialects diglossia editor ellison english ferguson finite generation geroge habash ibrahim imad information journal kareem kharashi kiraz languages level linguistics montereal morphological morphology newton nizar only operations owen page pages philadelpia phonology proceedings rambow references rosner science seminar semitic shallow society state sughaiyer survey techniques technology twolevel using version word workshop
http://acl.ldc.upenn.edu/P/P06/P06-1026.pdf	25	Learning the Structure of Task-driven Human-Human Dialogs	aaai across action acts adapted adjoining agenda agents alexandersson allen almost amounts amsterdam analyzing annotation application applying approach architecture artificial attention automated automatic automatically automation automatique autonomous baltimore bangalore based bear benjamins berger bohus build carberry carletta center charniak classifiers clauses coding cognitive collaborate collaborative collagen communication communicator computational computer computing conclusions conference constraints contact context continuous conversational conversations core corpora corpus correction corrective current damsl data datadriven decision decomposition deliverable description detection dialog dialogue different dipper discourse discoursep distributed down driven dysfluency edit editor entropy estimation eurospeech expectation experimental experiments extracting fabbrizio feature features first florence following formalisation fragments framework frampton from game general gilbert grammars grosz gruenstein gupta haffner hardy hastie henderson hierarchical hopkins human hybrid icslp identification identified identifying ieee ijcai information integrating intelligence intelligent intended intentional intentions interaction interfaces international interpretation interpreting into intonation introduction island jeju john johns johnson joshi journal jurafsky knowledge korea labeling language langues large larsson last learning lemon levin lewis linguistics litman lochbaum looked machine magazine management manager manaster manual margin markov mateo mathematics maximum meteer method mikheev model modeling models multiple multithreaded naacl natural necessary next njfun note observable observations october online optimizing order parsing partially patterns people performance pieraccini pietra plan planning poesio policies possible power practical predicting predictive prior probabalistic probabilistic proceedings processes processing project prosodic prosody pynadath ramer ravenclaw reasoning recognition references reflecting reinforcement reithinger relaxation reliability repairs report research response results rich roark robust rudnicky samuel scaling scheffler scheme science segmentation seneff sensitive sentences september sets should shriberg sidner sigdial signal simulation singh sources speech spoken spontaneous spring state statedependent stochastic stolcke strategies strategy structure structures stylebook subdialogs subtask such supertagging supervised switchboard symposium system systems tagging tags task taylor technical techniques topic topics traitement transactions transcribed transformation tree trindi trindikit uncertainty understanding university update user using utterance utterances views virtual wellman what when williams with workshop wright young
http://acl.ldc.upenn.edu/P/P06/P06-2075.pdf	221	Integrating Pattern-based and Distributional Similarity Methods for Lexical Entailment Acquisition	barcelona berland bernardo cafarella challenge challenges charniak chklovski coling corpora dagan distributional downey emnlp entailment etzioni eugene extraction feature finding fine geffet geneva glickman grained harris hypothesis inclusion information knowitall language large lexical maayan magnini maryland mathematical matthew michigan mining oren pantel parts pascal patrick popescu proc quality recognizing references relations scale semantic shaked similarity soderland southampton spain structures switzerland textual timothy vector verb verbocean very weld wiley workshop yates zelig
http://acl.ldc.upenn.edu/P/P06/P06-2121.pdf	267	HAL-based Cascaded Model for Variable-Length Semantic Pattern Induction from Psychiatry Web Resources	addison american baeza chen clinics information journal modern neto psychiatric psychiatry reading references retrieval ribeiro virtual wesley yates
http://acl.ldc.upenn.edu/P/P06/P06-2007.pdf	153	N Semantic Classes are Harder than Two	anick based bernardo challenge challenges dagan entailment feedback glickman magnini oren pages pascal peter recognising references refinement search sigir study terminological textual using workshop
http://acl.ldc.upenn.edu/P/P06/P06-2002.pdf	148	A Rote Extractor with Edit Distance-based Generalisation and Multi-corpora Precision Calculation	agichtein annotating artificial bases berland between brin charniak cimiano collections conference construct corpora craven database dipasquo edbt engineering extending extracting finding finkelstein freitag from global gravano handschuh icdl info infrastructure intelligence international knowledge landau large learning mccallum methods mitchell morin nigam ontologial pages parts patterns plain proceedings references relations relationships self semantic slattery snowball staab supervised technology terms text towards unsupervised very webdb wide workshop world
http://acl.ldc.upenn.edu/P/P06/P06-2077.pdf	223	Reinforcing English Countability Prediction with One Countability per Discourse Property	acknowledgments addition additional advantage advantages aldag almost also annual applicability applied appropriate araki article authors baldwin based bateson baxter bertolo bond bootstrapping british burnard cambridge church companion comparison compound computational computing conclude conclusions conference confidences context corpus count countability criteria curtis darpa data described detecting determine developed disambiguation discarded discourse distinction distinguishing earlier effective efficiently empirical english errors even eventually experiments exploiting explore extension first from further future gale generated give gives grammar guide handcoded hara have huddleston human improvement including increase inducing instances international intervention introducing investigate joint kawai keller kind language lapata learning lehmann lexical like linguistics long mappings mass masui means meeting method methods might mispredictions models most nagata national natural noun nouns ontology original other outperformed overrode oxford pages panton paper peng plethora prediction predictions press proc processing property proposed pullum recall reference references reinforce reinforcing requires result results retagged rivaling rules salay satoshi schneider second sekine semantics sense services shown smith some speech subsection successfully supervised system tagged tagging thank that their these they this training transactions university unsupervised used users using vatikiotis version volume wagner wakana what when which will witbrock with word wordnet work workshop would yarowsky
http://acl.ldc.upenn.edu/P/P06/P06-1005.pdf	4	Bootstrapping Path-Based Pronoun Resolution	acquisition anaphora annual aone approach artificial association automated automatic barbu based bean bennett bergsma canadian catalina cherry chinatsu colin computational conference contextual coreference david eighteenth ellen evaluating evaluation expectation gender information intelligence knowledge learning linguistics manual maximization meeting methods mitkov naacl pages proceedings pronoun references resolution riloff role rule ruslan scott shane strategies tool unsupervised william
http://acl.ldc.upenn.edu/P/P06/P06-1016.pdf	15	Modeling Commonality among Related Classes in Relation Extraction	advances anlp aone april arbor automatic bagging barcelona bartlett based breiman bunescu chunk classifiers clustering coling collins combining comparisions computer conference content corpus culotta database december dependency discovering dissertation driven edited emnlp entities entity entropy exploring extract extracting extraction features from grisman head http ijcnlp information integrated international island jeju jian journal july june kambhatla kaufmann kernel kernels knowledge korea language large learning lecture lexical lexicography likelihood lncs machine machines margin mateo maximum message methods michgan michigan miller models mooney morgan named natural nips notes novel online outputs parsing path pennsylvania philadelphia platt poster predictors press probabilistic proceedings projects ramshaw reasoning recognition references regularized relation relations research richardella roth scholkopf schuurmans science seattle semantic shortest similarity smola sorensen south spain statistical subsequence support syntactic tagger taiwan text tree understanding univ university upenn using vancouver vancover various vector wang weischedel with wordnet zelenko zhang zhao zhou
http://acl.ldc.upenn.edu/P/P06/P06-4011.pdf	297	Computational Analysis of Move Structures in Academic Abstracts	abstracts academic american analysis ansi anthony article aspects assist aston author barcelona bilingual birmingham cambridge chang collocational communication concordancer demo distribution english genre ieee institute introductions jian language lashkia learning machine medical meyer modality move mover national papers post press prof purposes reading references research salager settings specific standard standards studies study swales tango technical tense text tool trans type unit university verb writing york
http://acl.ldc.upenn.edu/P/P06/P06-2037.pdf	183	Low-cost Enrichment of Spanish WordNet with Automatically Translated Glosses: Combining General and Specialized Models	aaai aachen accompanying aclaracion acompana actual adjuntas agirre algo alicia alignment alon amta analysis andreas antonia approach atserias attained automatic banerjee barcelona based beam bernardo bibliograf bleu breve brian brief brown cada cambridge capullos carroll center central chin cionario cocke cocoons coling comparison computational computing conference cooccurrence corpora corpus correlation corresponds cosa csail cual database decoder della descripcion description desde diario diccionario dici doddington editor eduard electronic emnlp eneko espanola europarl evaluation every expected extensible external extrinsic factoid fellbaum fibers fibras flannery franz fredrick from generating george german gram gran green gusano hacer happens hermann hermjakob hilos hovy http human icslp illustration ilustracion improved improving input internation intrinsic jelinek john jordi josef joseph judgments kishore knitting knowledge koehn language larousse lavie lengua lexical linguistics luis machine magnini mart meaning measure measures melamed mercer meteor method metric mihalcea modeling models moldovan monotematica multilingual naacl newspaper numerical oute output outs pages papineni para parte paul people pequena periodica periodico peter pharaoh philipp phrase piek pietra planeta precision press proceedings proporcionan provide publica publicacion publications published punto quality rada rate recall recipes refer reference references report repository research respectively rigau ritmo robert roossin rosca roukos rwth ryan salim satanjeev saul scientific search seda sense significance silkworm single something source speech srilm statistical statistics stephan stephen stolcke sucede summarization superior systematic systems systran table tagged tapa tarifa task technical technology tejer templates tests teukolsky that thesis threads todd toolkit tospeech translation trec tribble turian university using various vetterling villarejo vincent vogel vossen ward watson weijing which william with word wordnet workshop
http://acl.ldc.upenn.edu/P/P06/P06-1046.pdf	45	Scaling Distributional Similarity to Large Corpora	algorithms andrei annual approaches approximation april association automatic best broder burkhard canada charikar communications complexity compression computing containment curran david distributional documents edinburgh estimation extraction file from goemans group improved improvements interest italy james journal july keller lexicon machinery marc match maximum michel moens montreal moses november pages philadelphia problems proceedings programming quebec references resemblance robert rounding salerno satisfiability searching semantic semidefinite sequences similarity some special symposium techniques theory thesaurus thesis university using walter williamson workshop
http://acl.ldc.upenn.edu/P/P06/P06-2114.pdf	260	Sinhala Grapheme-to-Phoneme Conversion and Rules for Schwa Epenthesis	academic account accurately adamson addressed akuru alan algorithm algorithms applications approaches arabic asanka association august available bali bangalore basaka based been being best beta black blue branch bros building carnegie cases cepstral choudhury cocosda coling colombo communication comparative comparison computational computing conclusion conference congress contemporary conversion corpus damper daniel dealing delhi demonstration developed development disanayaka documented domain dordrecht download downloads dutoit education empirical epenthesis esca evaluation evidence festvox focus from fully gamage general geneva godage grammar grapheme graphemeto gunasena gustafson helsinki hindi html http hussain ieice ijcnlp imam implemented india indian indicates institute integrated interna international introduction isca island issues james jeju joint jurafsky kalika karunatillake kevin kluwer knowledge korea krishna kularathna kumudu laboratory language languages lemmetty lenzo letenglish letter linguistics literary literature ltrl maharagama mahima malay mapping marchand martin mawatha mechanism mellon method model module monojit mountains national natural nearly netherlands olcott other pagel pages paper partha patparganj pearson phoneme phonetic pili pittsburgh platform pratim presented problem proc proceedings processing proposed provides publishers ramakishnan recognition references reported research retrieved review rule rules ruvan sami sarmad school schwa science script singapore sinhala site sound special spectrum speech spoken sridhar standard structure switzerland syllabification synthesis synthetic syst system talukdar techniques technologies technology text that there thesis third this tional tool tools tospeech tralia trans transcription ucsc university urdu useful various vincent voices wasala weerasinghe wide will with workshop yousif zuraidah
http://acl.ldc.upenn.edu/P/P06/P06-2053.pdf	199	Towards the Orwellian Nightmare	annotation bailando berkeley berkely brian carnegie classification coling communication corpus data david deceptive detecting diesner document elbert email enron exploration from guthrie html http introducing jana karley kathleen keila klimt louise machine mellon networks parambir practice proc queen references research sims skillcorn theory university unusual walker yang yiming
http://acl.ldc.upenn.edu/P/P06/P06-1122.pdf	121	Modelling lexical redundancy for machine translation	acknowledgements additional addressed adjectives alex alignment also amta analysis annotated annotation area association authors backoff based basu bayesian beam besag bilenko bilingual brown case categorised challenging chapter characterise chou classes classification cluster clustering cmejrek comments comparison computation computational conclusions conference context corpora corpus could curin czech daniel data david decoder define defining della demonstrated dependency determining dirty discovery distinctions distribution distributions edinburgh efficient emnlp empirical english entropy estimation europarl european evaluation experiments feature features framework franz from future gains geman gibbs goldwater group hajic havelka hermann highly however human icslp ideally identify ieee images improvements improving inclusion incorporating inflected informatics information instance intelligence international investigate john josef journal julian katrin kirchhoff knowledge koehn kubon lafferty language languages lapata learn lexical lexicon like linguistic linguistically linguistics lisbon machine marcu mathematics maximum mcclosky members mercer method methods mikhail mining mirella model modelling models monolingual mooney more morpho morphological naacl natural niessen nouns number only optimal optimisation over parallel parameter partitioning pattern peter pharaoh philadelphia philip philipp phrase phrasebased phrases pictures pietra portugal prague preliminary prior priority probabilistic proc proceedings processing proposed providing quality raymond redundancy redundant references regression relaxation renals resources respect restoration resulted robert royal same scarce school scripts search selection semi series sets sharon showed sigkdd significantly society spoken statistical stephen steve stochastic structure studentship sugato suggest summit supervised supported syntactic syntactically systematic sytem tackled technology thank that then these this through training trans translation treebank trees university used using valuable various vincent waibel wang weights whether while will with within wolpert word work workshop would yang zens
http://acl.ldc.upenn.edu/P/P06/P06-2101.pdf	247	Minimum Risk Annealing for Training Log-Linear Models	accuracy achieve achieved after algorithm algorithms annealed annealing another applying approach approximating arbitrary attempt audio automatic away bahl based bayes best better both bottleneck brown bulgarian byrne calculate carnegie cases challenging charniak chen classification classifier clustering coarse cohn collins community comp compression computer conclusions conditional conll convex corpora crammer data decoder decoding densest dependency dept descent design despite deterministic deterministically different discriminative distinction does dreyer dutch dynamic each eisner either elidan emnlp entropy error estimates estimation evaluation experts family fields fine fitting found friedman from function functions gains gaussian generalized global goel goodman have heldout helped hemisphere hidden hinton hoffgen hope horn however hypothesis icann icassp icml ieee improved incorporates indistinguishable inferior information inside interested jmlr johnson juang just katagiri klein koehn koller kumar labeled labeling lafferty language large learning like likelihood line linear literature logarithmic loss lower machine make manner manning marcu margin margins markov maxent maximum mccallum mcdonald measures mellon mercer method methods metrics minimization minimizing minimum model models more naacl nakano networks neural neurons never nips nonlinear note november numerically ones opinion optimization optimize optimizing orthogonal osborne other outputs over pages papineni parameters parsing pattern pereira performing performs phrasebased pools practical precision prediction preparata previous prior probabilistic probability problem problems proc procedures processing produces products programming random rate rather recognition recognizers references regression related reminiscent report required reranking researchers results risk robust rose rosenfeld roukos sciences search seen segmenting sentence sentences sequence shape showed significant significantly simon single slovenian smith smooth smoothing some sometimes souza specific speech speed statistical structured such surface system systems table task taskar technical technique techniques test than that their theoretical these this three trainability training trans translation ueda university unsupervised upon used using variable vine volume ward weights while with worse
http://acl.ldc.upenn.edu/P/P06/P06-2034.pdf	180	Discriminative Reranking for Semantic Parsing	algorithms annual arbor association available best boosting brooke canada carreras charniak chen coarseto collins computational conf conll cowan discrete discriminative ehsan emnlp empirical entity eugene experiments extraction fine foroughi fredrik generative heintz hidden http human icml integrates intl introduci itsuki johan johnson july june kapetanakis kernels kostas kostiadis kummeneje labeling language later learning lexicalised linguistics machine manual mark markov marquez maxent meeting methods michael models mooney morphology named natural noda obst october oliver over pages parser parsing patrick perceptron philadelphia proc processing projects ranking raymond references reranking riley robocup role ruifang semantic semantics server shared soccer sourceforge spanish spiros sserver stanford statistical steffens structures syntax tagging task technology that theory three timo tion training users vancouver version voted wang with xavier xiang
http://acl.ldc.upenn.edu/P/P06/P06-1079.pdf	78	Exploiting Syntactic Patterns as Clues in Zero-Anaphora Resolution	acyclic adaptive alexandrov algorithm analyzing anaphora anaphoricity antecedent applications approach approaches asahara asian assignments automatic automatically baldwin based boosting cardie carreras case centering classification cleaning cogniac coherence coling collins communications competition computational computer conditionally conll consolidation constraint contextual control convolution coreference cues data decision definite deictic department description detection detectors determination dialogue dictionary different directed discourse distance duffy eacl electronic emnlp engine factors first followed framework fujii fujita generated gildea goulart graph grosz guide hayashi hierarchical hirao identification iida ijcai ijcnlp ikehara improve improving incorporating incorrect information institute integrating introduction inui ipadic ishikawa issues iwanami japan japanese john joshi jurafsky kabadjov kameyama kernel kernels kudo labeling language lappin learning leass lehnert linguistics linkage local machine maeda manual marquez matsumoto matter mccallum mccarthy method methods metric mitkov miyazaki modeling models muller nakaiwa nara natural nihongo ninth nips noun object ogura okumura only ooyama operational optimization pages paraphrases partitioning pennsylvania phrase phrases poesio practical preliminary probabilistic proceeding proceedings processing pronominal pronoun pronouns property proposal record reference references representation research resolution robust role roles sasaki science sciences seki semantic semi sentences shared sharing shirai shoten signal sons soon spoken statistical strube structured study survey suzuki systems taikei takamura talip tamura task technical technology text that theory thesis they things trainable trained transactions treatment trees university uryupina user using vapnik vieira weinstein wellner wiley with workshop yang yokoo zero zhou
http://acl.ldc.upenn.edu/P/P06/P06-4010.pdf	296	Chinese Named Entity and Relation Identification System	abney academic acknowledgement acquisition added also application applications approach appropriate atomic automata automatic automatically based beijing beneficial bosch boundary brill calculation cannot cascades case cases challenge chen china chineris chiners chinese cken collate commercial complex component components computational conclusion constraint constructing constructons contract correcting czech daelemans definition determination develop dfki ding distance domain dong driven each editor editors education effective effectual engineering enhances entity erbach error errors esslli experimental exploration extend extension extraction feature features finite first flexibility german germany guide hommel hownet html http identification important improve increases informatics information innovative internet issue japan joint jscl keenage kluwer knowledge language learner learning level lexical life linguistic linguistics machine maintenance mechanism medical medicine memory michigan ministry multi named nanjing national natural ners netherlands novel ontology opposite order original pages papers parsing part partial performance plays pncbl points practicability prague press proc proceedings processing project prot protege prototype publishers rather real recognition recognized reference references relation relations repairer report republic research results robust role saarbr same sample sapporo school second segmentation selecting semantic sheng show sighan similarity simultaneously sloot some special speech sports stage stanford state strategies study successful such supported symbols symposium system tagging technical technology text than that theoretically third this three tilburg timbl time transformation trigger types under university uszkoreit vanden version which with without word words work workshop zavrel zhiwang
http://acl.ldc.upenn.edu/P/P06/P06-1020.pdf	19	Morphology-Syntax Interface for Turkish LFG	abeille academic affixation alternations alternative andy anette anne annie annotated annotation aoife appear approach arbor association august automatic baitin bank baris barker bergen bilge bozsahin bresnan building burke butt cahill cakici cambridge categorial causative cfgs chapter chicago chris combinatory computational computing conceptions conference constituent corpora csli dalrymple dependencies dependency description dilek donovan dordrecht eacl editor editors efficient elisabeth engdahl eryigit european evaluation exploiting extended extracted finite formal formalism frank from functional genabith germany gokhan grammar grammatical grenoble guistics gulsen gungordu hakkani hankamer holloway hpsg induction international italy joan john joint jorge josef june kabak kaplan kemal king kluwer kroch large level lexical lexicalfunctional lexicon linguistic linguistics literary longdistance louisa machine mary maxwell mental michael michigan miriam moore morphemic morphological morphology norway oflazer pages parser parsing penn phrase press proceedings publications publishers rank references rela relations relativization representation research resources restriction ronald ruken ruth saarbrucken sadler scale semantics state statisti structure structures student suspended syntactically syntax system tional tracey tracy translation tree treebank treebanks trees trento turkish uncertainty university urdu using valency volume with workshop xerox york zaenen zelal zeynep
http://acl.ldc.upenn.edu/P/P06/P06-2035.pdf	181	Multilingual Lexical Database Generation from parallel texts in 20 European languages	acuerdo adelaide agreement alignement aligning alignment alignments altenberg amended amendments anexo anhang annex annexe annual apidianaki approv approval approximat article artikel association authorities automatique autres avis barriers berkley between bilingual borovets boutsis brown bulgaria cada california cation certain certificat certificate changes ches church citations clauses comisi commercialisation commission communities community comp competent computational comuni concept condiciones conditions conference conrast conseil consejo consid considerando corpora corpus correspondences council cross crossing culo dades dans darpa deux device dictamen directhis directithis directiv directiva directive discours discovery dispositif documents dure each economic ello empirical endogenous english estado estados europ european europeas font force fourth french freq fresh from fronti gale gemeinsamen gemeinschaft geof german gest giguet grained granada granger grounding harmattan having hectolitre identifying informatifs initial insbesondere internationales into ischen jean journ juillet july june juni junio kaufmann kommission language learning lexical lexis linguistic linguistics linguistique liquid liquides locuteurs lorient lucas mark market marks marnette marque marques mass masse mateo material measures meeting meinschaften member membre membres mento mercado mercer meses methods miembro miembros mitgliedstaaten modifications months morgan morphemes multi multiplication nach nahmen natural ndert ndoc necessary nicht objet oder only opinion order other pages pagnll paradigms parallel paris particudirective particular pattern piperidis poids poultrymeat presente primi primitive proc procedure processing program proposal proval question ranlp rant rapport rapprochement rden recent references regard reglamento reglathis regulation relat relativ research resources richtlinie rification rosier sente sentences single some source spain spanish speech state states stellungnahme structures studies subject summarization symbol taille target tats technical technischen tection tent text textes texts textuelles than that thereof third this tiva tive todos tous trade tratado treaty trends type unit untagged variable vehicle verification verifiv verordnung vertrages viandes vigueur visto volaille vorschlag votes votos weight whereas wirtschaftsgemeinschaft with word workshop zwischen
http://acl.ldc.upenn.edu/P/P06/P06-4006.pdf	292	MIMA Search: A Structuring Knowledge System towards Innovation for Engineering Education	ananiadou application approach articles atract automatic biological blast cambridge ciee clustering coling copenhagen denmark dialogue discovering evaluation extraction friedman from fukuda gene hawaii hierarchical http identifying information jacquemin japanese journal krauthammer lnai matousek mautner mima morozov moucek multi names nenadic papers press proc protein recognition references rzhetsky speech spotting springer takagi tamura tauser term terminology terms text through tokyo toward tsunoda units ushioda using value verlag word words workbench
http://acl.ldc.upenn.edu/P/P06/P06-1006.pdf	5	Kernel-Based Pronoun Resolution with Structured Syntactic Knowledge	achieves acquisition against algorithms american anaphora annual aone association automated bennett brackets chapter charniak collins compuational computational crossing discrete duffy entropy evaluating inspired kernels linguistics manual maximum meeting north over pages parser parsing perceptron precision proceedings pulic ranking recall references reports resolution section strategies structures tagging their treebank voted with words
http://acl.ldc.upenn.edu/P/P06/P06-1096.pdf	95	An End-to-End Discriminative Approach to Machine Translation	achieve acknowledgments adding aggressive algorithm algorithms allows also amongst andreas anoop approach approaches approximate ariadna automatic based baseline best better bleu bounds brian brown candidate categories chiang chooses christoph class classifier collins comments common components computational concern conclude conclusion conditional conference conservative contrast corpus currently daniel darrell david decision decisions decoder decomposed della dependence desouza discriminative effect emnlp entropy error estimation europarl evaluation experiments explicitly expressive extensible extrinsic fall fashion feature features fields first formulation franz from gains general generates generative gildea global gram handful have helmut hermann hidden hierarchical hltnaacl icslp important improvement improvements increase incremental instance international into intrinsic investigated issue iteration iterations jennifer john johnson josef khudanpur koehn language large learned libin like limitation linguistics list lists local localized machine major making many marcu mark markov mathematics maximum maxwell mercer methods michael minimum model modeling models moderately monotonic most mtse multilingual murat naacl natural nips novel number object online only optimize over parameter parameters parsing part perceptron performance peter pharaoh philipp phrase pietra pitfalls poor potential practice prediction present presented previous probabilistic processing quality quattoni random randomization rate real recognition references related relative reported required reranking result reuse reviewers richard riezler roark robert sanjeev saraclar sarkar schmid score second separate sequence shen significance significant sized small smaller smorgasbord some speech srilm statistical stefan stephen stolcke strategies studied summarization suponly system tagging terry testing than thank that their them theory this tillmann tong toolkit training translation translations trees trevor tune underlying update updates used using value variable vincent weights which while with work works workshop would yielded zens zhang
http://acl.ldc.upenn.edu/P/P06/P06-2062.pdf	208	GF Parallel Resource Grammars and Russian	aarne appear approach arabic articles based binding blackwell bouillon bringert burke butt cambridge caprotti carter chalmers chicago chomsky comprehensive computation cookbook cooper csli dada deliver dialogue dialor dictionary digalakis dordrecht driven editors eldada engine engineering everywhere foris formal formalism forsberg framework france french functional government grammar grammars grammatical grammatike head holland icfp implementing italian johannisson journal kalamazoo kellogg king lacl language lectures line linguistics ljunglof lnai mathematics michigan modular morph morpho morphology moscow multieng multimodal nancy natural online orlando pages paper phrase pisa pollard press programming publications publishing ranta rayner references research resource rinet russian russkoj russky segond shelyakin site software spanish specifications spoken spravochnik springer stanford starling starostin structure stud symposium system translating translator typetheoretical university wade webalt wiren wordreference writer yazyk
http://acl.ldc.upenn.edu/P/P06/P06-3008.pdf	279	Discursive Usage of Six Chinese Punctuation Marks	aaai academic aligning automatic bilingual building carlson chan chinese chuang computational computing corpora corpus corston criteria current dale daniel dialogue directions discourse editors framework kevin kluwer kuppevelt language linguistics lynn manual march marcu markers mary mining notes okurowski oliver parallel processing publishers punctuation reference references report representation rhetorical robert role ronnie samuel simon smith statistically structure summarization tagged tagging technical textual theory thomas with working workshop written
http://acl.ldc.upenn.edu/P/P06/P06-1031.pdf	30	A Feedback-Augmented Method for Detecting Errors in the Writing of Learners of English	allan america bond british burnard chapter chodorow computing corpus countability csli detecting dictionary electronic errors grammatical guide institute japan leacock linguistic meeting method national north nouns oxford pages proc publications reference references research services society specifications stanford translating university unsupervised untranslatable users version
http://acl.ldc.upenn.edu/P/P06/P06-1111.pdf	110	Prototype-Driven Grammar Induction	abney access achieve after alexander algorithm allow also andrew annotated another applications approach approaches arnold aspect association authors automatically auxiliary based beatrice being best bracketed building built cambridge carroll categories charniak chicago chinese christopher clark clustering collins compact computational computer conclusion conll constituency constituent context contrastive corpora corpus data declarative degraded demonstrate dependency described distribution distributional eacl edition edward eisner english error estimation eugene experiment experiments expertise extracted features fernando figure first foundations free from functional gave generative give glenn grammar grammars grammatical guiding halliday harris have highest hinrich ijcai importantly improved increased independent inducing induction inference insideoutside introduction ircs jason karim klein knowl labeled labeling language large lari learning length linguistics lisbon literature manning marcinkiewicz marcus mary measure meeting michael mitchell model models more most natural next nianwen noah notes noting noun only order pages part partially penn pereira performance phrase portugal positive press presumably primarily probabilistic processing property prototype prototypes punctuation radford reconcile reduction reestimation references relative report reported respectively results rochester roto same santorini scale schabes scheme schutze section sentences sentential shown similarities since smith somewhat speaker specify speech stanford statistical stephen steve stochastic stripped structure substantial success supervised syntactic system tagged tagging target technical tested text that these thesis they this transformational tree treebank tries university unlabeled unsupervised using with work working workshop worth would young yves zellig
http://acl.ldc.upenn.edu/P/P06/P06-1052.pdf	51	An Improved Redundancy Elimination Algorithm for Underspecified Representations	about achieves algorithm algorithms also althaus ambiguities ambiguity analysis appear applications approximates arbor berlin between blackburn brants bridging building certain chart chaves claims class coling collaborative complete completeness computation computational conclusion constraint constraints copestake corpus could course csli currently deemter defined definition deletes demonstration described descriptions difficult directions disambiguation does dominance drops duchier each eacl earlier editors efficient eliminable eliminates elimination engineering enumerating enumeration equivalence esslli evaluated evaluation even evolution exploiting exploration explore factor fifth first fits flickinger formalisms formulas from fuchss further generating grammar graph heuristics hole icos ideas improve improved improvement improves individual inference information intl introduction irredundant journal koller lambda language least lingo literature logic made maintains making manning matter maximum median mehlhorn minimal more most motivation natural negligible never niehren note notes number oepen original over papers peters pollard polynomial practice preliminary present presented previous proc proceedings publications pursuing quantifier readings recursion reduced reduces reduction redundancy redundant redwoods references remains representation representations representative respect rewriting rondane runs runtime scope scopings seconds semantic semantics sentences session setting shieber showed size software solvers solving some splits stanford structures student successively sure systems take that thater then there these thiel this time toutanova towards translation treebank tsujii types underspecification underspecified underspecifiedness upon useful uszkoreit vestre well whereas which while with work workshop worthwhile would year
http://acl.ldc.upenn.edu/P/P06/P06-2001.pdf	147	Using Machine Learning Techniques to Build a Comma Checker for Basque	account aduriz akman aldezabal algorithm analyser annual aranzabe aren argitalpen arrieta arriola arti aspects atala automata basque berlin brunswick buchholz california cascaded center checker checking chunking clause combining comma commas computa computational conference conll copenhagen corpus country cruz cuny danish delden denmark determine duction england estilo eusko ference fication ficial finite france gasteiz germany gojenola gomez greedy hardt hill human identi ieee ilarraza information informationbased intelligence intelligent international intro introduc jaurlaritzaren jean jersey jones junior lancaster language learning lehen leland liburu linguistics lisbon lncs maritxalar meeting murray nagusia nunberg oronoz ortotipografia pages point portugal proceedings processing punc punctuation references roles sang santa sentence series shared sigparse spaces spain springer stanford state study syntactic task ternational text tion tional tjong tolouse tools toward towards tuation university uria verlag washington with zerbitzu zubimendi
http://acl.ldc.upenn.edu/P/P06/P06-1105.pdf	104	Japanese Dependency Parsing Using Co-occurrence Information and a Combination of Case Elements	advances aleksandr andrew annual appelt argument automatic best charniak christopher coarseto collins computational conference daniel data defined derived development discriminative douglas eugene exact factored fast fine frequencies from gildea henderson hofmann indexing inference information international interpretation ivan james johnson jurafsky kehler kernels klein labeling language lara latent linguistics manning mark maxent meeting michael model models naacl natural neural nips pages parse parsing predicate probabilistic proceedings processing pronoun references reranking research retrieval roles semantic sigir simma systems taylor terry thomas titov utility with
http://acl.ldc.upenn.edu/P/P06/P06-2079.pdf	225	Examining the Role of Linguistic Knowledge Sources in the Automatic Identification and Classification of Reviews	aaai about adjectives advances affect analysis analyzer annotations answering applied appraisal argamon based bell bruce bunescu cardie categorization cikm classification coling collier comparative computational conference contextual corston customer cuts data dave dependency determining directions diverse documents down eacl education emnlp esuli etzioni evaluation examples exploiting extracting extraction facts feature features free from fukushima gallery gamon garg given gloss groups hatzivassiloglou hltemnlp hoffmann hovy icdm icml identifying ieee ijcai improve independent information intelligent interfaces international joachims kernel knowledge koppel labeled language large lawrence learning level lieberman linguistics litman lowlevel lrec machine machines main making martin mccallum mckeown methods minimum mining minipar mitchell model models morinaga mullen multi naacl nasukawa natural neutral niblack nigam oliver opinion opinions orientation pages pang parsing peanut pedersen peng pennock perspective phillips phrase polarity popescu poster practical predicting press proc processing product pulse question questions real recognizing references representations reputations reviews riloff ringger scale schler schuurmans sebastiani selection selker semantic sensing sentences sentiment sentimental sentiments separating simple sources study subjective subjectivity summarization summarizing summary support symposium systems task tateishi techniques terms text textual through thrun thumbs topic towards turney unlabeled unsupervised user using vaithyanathan vector wang whitelaw wiebe wilson with workshop world yamanishi yang
http://acl.ldc.upenn.edu/P/P06/P06-2003.pdf	149	MT Evaluation: Human-like vs. Human Acceptable	akiba amigo annual anselmo assistance assisting association automatic based booked campaign clock compartment computational costa could crego doddington endo enrique evaluation federico felisa fonol framework george getting give gonzalo hand have hello help helping high hiromi human ichi international into iwslt japan judgements julio june jussa kando kyoto language like linguistics losa machine made marcello marino meeting michael michigan mind nakaiwa name ngram nine noriko overhead overview pages paul penas phrasebased please proceedings putting qarla queen rack reference references reservation reserved score seats some spoken statistical storage sumarization table technology this translation translations tsujii under value verdejo versus with workshop would yasuhiro
http://acl.ldc.upenn.edu/P/P06/P06-4020.pdf	306	The Second Release of the RASP System	accuracy accurate annotation applied australia baum briscoe cambridge canaria carroll coling computer conf conference depbank does elworthy estimation evaluating evaluation general germany grammars gran help introduction laboratory language lrec palmas parc parser proceedings rasp references report resources robust sequence statistical stuttgart sydney system taggers technical text university unlexicalized welch
http://acl.ldc.upenn.edu/P/P06/P06-2061.pdf	207	Integration of Speech to Computer-Assisted Translation Using Finite-State Automata	acoustics aided alignment amta annual approaches april arbor asru assisted assoc association august automata automatic barcelona based bender beyerlein beyond bisani bootstrap boston brousseau brown budapest building canada chapter chen coling combination communication computation computational computer conf conference confidence copenhagen data della demand denmark denver devices dictation direct discriminative document driven drouin dymetman eacl editors efficient enhanced entropy error estimates estimation europ european eurospeech evaluationx extensible farwell feature finite flexible foster french fugen gerber greece hasan hmmbased hovy human hungary icassp icslp ieee improvements interactive international interspeech intervals isabelle istanbul iwslt japan juan july june kanthak kehler khadivi knight kuhn language large lecture likelihood linguistics lisbon machine madrid mathematics matusov maximum mediated meeting mercer michigan minimum model modeling models molau montreal naacl normandin notes novel october onaizan pages papineni parallel parameter paulik performance philadelphia phrase phrasebased pietra pittsburgh plamondon portugal proc proceedings processing project puerto rate recent recognition references reordering rhodes rico roukos rwth sapporo schaaf schluter schultz science search seattle september signal sixtus spain speech spoken spontaneous springer srilm state statistical stolcke stuker system targettext technology text texts tillmann toolkit towards training translation translators transtalk turkey understanding using verlag vilar vocabulary vogel volume waibel ward with word workshop yokohama zens zhang zolnay
http://acl.ldc.upenn.edu/P/P06/P06-1011.pdf	10	Extracting Parallel Sub-Sentential Fragments from Non-Parallel Corpora	accurate adaptive algorithm aligning alignment among appear approach articles artificial automatic backoff bilingual bing bleu bootstrapping brown byrne callan cheung coincidence coling collection comparable comparison computational conf conference constraints corpora corpus cross cyril daniel data dejean dekai della deng diab discovery douglas dragos dunning eacl emnlp engineering english entity equivalence eric estimating estimation evaluation events experiments exploiting extracting extraction finch franz from fung gaussier geometric german gina goutte grammar hermann herve hitoshi hwee identification ieee ijcnlp improved improving intelligence inversion irina isahara japanese jean joseph journal kevin kishore knight koehn kumar language lemur lewow lexicon likelihood linguistics lrec machine marcu masao mathematics matveeva measures melamed mercer method methods michel mining model models mona monolingual moore multilevel munteanu named national natural news noah noisy nonparallel oard ogilvie pages papineni parallel parameter pascale percy performance peter philip philipp pietra probabilities quasi quasicomparable rapp rare ratios references reinhard reliable renders resnik retrieval riao robert roukos salim satoshi segmentation sekine sentence sentences shankar shao shinyama significance smith statistical statistics stefan stephan stephen steve surprise systematic template tests text texts todd toolkit transduction translating translation translational translations trec unrelated using utiyama various very view vincent vogel ward weijing william word wordalignment wordlevel words workshop yonggang yuen yusuke zhao
http://acl.ldc.upenn.edu/P/P06/P06-1083.pdf	82	Extracting loanwords from Mongolian corpora and producing a Japanese-Mongolian bilingual dictionary	accuracy also analysis annual approach association atsushi automatic average back based bayarmaa beaulieu bilingual chasen cheung choi collocations compared computation computational computer conclusion conference constraints corpora corpus correct corresponded cost cyrillic daigakushorin development devised dictionaries dictionary different does dual efficiency effort ehara enkhbayar entity evaluate evaluating evaluation experiments extract extracted extracting extraction figure finding first foreign frank from fujii fung gatford generation grades gram grammar hancock hatzivassiloglou hayata hidden hooh huang hyeok hyun improved improvement information international ishikawa japanese jeong jones jong journal katakana kathleen kimura korean language languages large learning lexicons linguistics loanwords maintaining management markov matching mckeown meeting memory method methods micheline mike mining model modern mongolian morphological myaeng named natural nist nobuyuki noun okapi oriental ozawa parallel pascal pentium performed phonetic phonetically phonological phrase proceedings processing produced producing proposed publication randomly rank references require research retrieval robertson ruizhang rules same sanduijav sato satoshi selected serious shan shigeo shown shows sigir similar similarity smadja soon sorting special statistical stemming stephen steve sung susan suzushi table takehito targeting term terminology terumasa tetsuya text that their them third time translating translations transliterated transliteration trec used using utsuro vasileios very walker were while words workshop
http://acl.ldc.upenn.edu/P/P06/P06-1127.pdf	126	Novel Association Measures Using Web Search with Double Checking	bagga baldwin based coreferencing crossdocument entity references space using vector
http://acl.ldc.upenn.edu/P/P06/P06-1119.pdf	118	Leveraging Reusability: Cost-effective Lexical Acquisition for Large-scale Ontology Translation	academy acquisition afita agricultural agriculture annotated applications araki artificial asian automatic beijing bilingual brown caas china chinese chun cmejrek computational conference corpora cross czech deep della dependecy echizen english enrichment estimation evaluation extraction federation from gaussier hajic havelka information intelligence international jean kubon language lexical linguistics lisbon machine mathematics medical medicine mercer momouchi multilingual pairs parallel parameter pietra portugal prague proceedings processing references renders resources retrieval rules sadat sciences siglex statistical syntactically technology terminology thesaurus third translation treebank wenlin word workshop
http://acl.ldc.upenn.edu/P/P06/P06-1061.pdf	60	Segment-based Hidden Markov Models for Information Extraction	aaai anlp bikel extraction finder freitag highperformance hmms information learning machine mccallum miller name nymble pages proceedings references schwartz shrinkage weischedel with workshop
http://acl.ldc.upenn.edu/P/P06/P06-1109.pdf	108	An All-Subtrees Approach to Unsupervised Parsing	algorithm algorithms based beijing best better beyond bonnema budapest cambridge chiang chicago clark clustering combining computational conll context csli data dempster distributed distributional duffy eacl efficient estimation experience free from geman goodman grammar grammars huang icslp implementation incomplete induction interpretation iwpt journal kernels laird language linguistics madrid maximum model modeling oriented over parsing perceptron philadelphia press probabilistic proceedings publications ranking references royal rubin scha semantic sima society statistical stochastic structure structures syntactic tagging theory university unsupervised using vancouver voted with york
http://acl.ldc.upenn.edu/P/P06/P06-1044.pdf	43	Automatic Classification of Verbs in Biomedical Texts	accurate allerton ambiguity ananiadou anaphoric annotation annual applied argument articles automatic based bialek biocomputing bioinformatics biomedical bootstrapping bottleneck brew briscoe canaria carroll case channels class classification clustering coding communication computation computational computing conference construction control corpora database decoding description dictionary dimitrov distributions dorr drosophila empirical entitites evaluation extraction foreign friedman from gasperin general german germany gran harris informatics information interlingual international journal language large lewin lexical lexicography lexicon line linguistics linking machine masterclass merlo method methods miller named natural network neural pacific pages palmas pereira philadelphia prescher probabilistic proc processing quantization reasoning recognition references resolution resources riezler robust rooth rzhetsky saarbrucken scale schulte spasic spectral statistical stevenson structure subcategorization sublanguages symposium system systems terms text theories tishby translation tsujii tutoring using verb verbs vlachos walde washington wordnet zellig
http://acl.ldc.upenn.edu/P/P06/P06-1090.pdf	89	A Clustered Global Phrase Reordering Model for Statistical Machine Translation	alex alicia american annual ashish association based bing bleu campaign chapter chiori coling computational conference constraints daniel eiichiro emnlp empirical evaluation figure franz global hermann hori huang human international iwslt joint josef koehn language linguistics local machine marcu matthias meeting methods model natural north overview pages philipp phrase proceedings processing references reordering richard score spoken statistical stephan sumita summit system taro technologies translation tribble venugopal vogel waibel watanabe workshop ying zens zhang zhao
http://acl.ldc.upenn.edu/P/P06/P06-2064.pdf	210	Interpreting Semantic Relations in Noun Compounds via Verb Semantics	acapulco artificial automatic baldwin banerjee barcelona barker canada compound computational conference eighteenth expressions extended getting gloss integrating intelligence international joint linguistics machine measure mexico modifier multiword nominals noun overlaps pages pedersen proceedings processing quebec recognition references relatedness relationships right satanjeev semantic semi spain stan szpakowicz takaaki tanaka timothy translation workshop
http://acl.ldc.upenn.edu/P/P06/P06-1023.pdf	22	Trace Prediction and Recovery With Unlexicalized PCFGs and Slash Features	accuracies accuracy accurate added advp again algorithm almost also although ambiguous american amit anlp annotated annual antecedent antecedents applies apply approximation association barcelona beam beatrice because best better bracketing building campbell categories category chapter charniak christopher coindexation coling collins combining comparison complementizers computational computes conference context corpus correcting data deep dependencies dependency dienes different dubey efficient elements empirical empty english entropyinspired eugene evaluated exceeding experiments extracted feature features fillers france free from general generate generative geneva grammar grammars guaranteed helmut highest highly improve indexation input international japan johnson june klein knowledge labeled language large levy lexicalised like likely linguistic linguistics lower madrid manning marcinkiewicz marcus mark mary matching maximum mechanism meeting methods michael mitchell models most naacl natural nodes north order other outperforming outperforms pages paper parse parser parsers parses parsing pattern pcfg pcfgs penn percolation perfect prediction presented previously principles proceedings processing produce pruning ranks reaching recover recovering recovery references reported representations richard roger santorini sapporo sbar schmid score search seattle shallow simple slash slightly spain standard statistical strategies summary surface switzerland syntactic system table tagger task terms test than that their they this three toulouse trace traces training tree treebank trees true unlexicalized unlike used using vectors version viterbi volume washington were which with
http://acl.ldc.upenn.edu/P/P06/P06-1121.pdf	120	Scalable Inference and Training of Context-Rich Syntactic Translation Models	alignment approach based bilingual binarization chiang cohesion coling computational corpora data emnlp galley gildea graehl grammars hierarchical hopkins huang inversion knight linguistics machine marcu model naacl oriented pages parallel parsing phrasal phrase poutsma proc references rule statistical stochastic synchronous syntax template training transducers transduction translation tree what yamada zhang
http://acl.ldc.upenn.edu/P/P06/P06-3015.pdf	286	Clavius: Bi-Directional Parsing for Generic Multimodal Interaction	about acknowledgements active ageno alexandersson algorithm already also analysis angeles annual apparently applying architecture artificial automatic avenues bangalore based basic becker been begun bidirectional bimanual bolt bourse boussemart brno broadcast chart chicago coanalysis command complex computational concurrent conf conference consistently constituents constraintbased context contextsensitivity cooperstock csli currently czech decide deduction deictic dialogue directed discourse discriminative discussed distribution dowding dynamically dysfluencies early edges efficient emphasised employ environments especially evaluation events expanded exploration explore extending figure filled finally finite first fonds form four framework free function funding further fusion future gemini gestural gesture gestures given grammar graphics hand have head holzapfel hong icmi ignores ijcai immersive implementation including incorporate incorporating intelligence interface interfaces intl itself johnston joint kettebekov knowledge kong language languages later lead least lends length level linguistics lopes lrec maitrisse manipulation mcneill meeting methods mind model modelling modules modulespecific more morristown multimodal narration natural nature near nickel ninth noise noisy nouns number observation ongoing open operation over overlay palmas parallelism parameters parser parses parsing partial particular party permitting pointing practical press proc processing prosody provided publications qualifications quebecois reasoning recherche references relevant remarks republic resources result reveal rioux rocio rodriguez rudzicz scoring seattle second sentence shared siggraph significant sizes space spain specifically speech spoken state stiefelhagen stochastic subspaces suggests system systems tabling tapd technologies that there thought tomita tracking train training understanding unification units university unordered unquantified untethered using verbs visualisation voice vrst weather what whether which will with words work working workshop wozniewski york
http://acl.ldc.upenn.edu/P/P06/P06-1045.pdf	44	Selection of Effective Contextual Information for Automatic Synonym Acquisition	accurate acquisition adopted also although american analysis annotated annotation annual applicable application argument association automatic baroni beatrice because better between beyond bigger bisi briscoe building carroll categories chosen christiane classification clustering cobuild coling collins combination common complexity computational conclusion conference considering construction contents contextual contribute contribution cooccurrence copestake corpora corpus database deerwester degree dekang dependency development different discover distributional diverse donald edition effect electronic english evaluated evaluation even examined existence existing experimental experimentally extent extracting fellbaum forum fourth framework from future general gives graham greater greatest hagiwara harpercollins harris hindle hirokazu hofmann ijcnlp impact including indexing information international investigated iwanami jerrold john joint jonathan journal katsuhiko katz kawaoka kazuhide kinds kojima language large latent lexical limited linguistics lrec major makoto marcinkiewicz marco marcus mary masato measures meeting might mitchell modification more nagao natural noun nouns object occurrence ogawa other overall oxford parseval penn perform performance performing philosophy plsi poorly possibility predicate press probabilistic proc processing proximity publishers radius references relational relations relationship relatively research resources result retrieval robust sabrina same santorini schemes science scott second selection semantic sentence series shizengengoshori shoten should showed shown shows sigir similar size society software some specific stability stable statistical statistics structure structures study subject synonyms target technical technology text than that their themselves thesaurus third this thomas three threshold toyama tradeoff treebank tsukasa university used using utilization watabe well while widely will window with wordnet words work workshop yasuhiro zellig
http://acl.ldc.upenn.edu/P/P06/P06-2029.pdf	175	The Benefit of Stochastic PP Attachment to a Rule-Based Parser	ambiguity analysis annual anytime approach arbor association attachment backed based berlin brill brooks christiansen coling collins combining computational conf constraint constraints corpora data daum defeasible disambiguation driven dubey editors fails foth german head hindle hybrid iwpt japan jersey journal kokkinakis kyoto language large lexical lexicalization linguistics lnai meeting menzel model models natural nordic pages parser parsing pennsylvania philadephia phrase predictors prepositional probabilistic proc processing properties references relations resnik rooth rule schroder skadhauge smoothing solving somerset springer statistical structural suffix supervised swedish sydney symbolic technique technologies text thesis through training transformation university unrestricted unsupervised using verlag very villadsen volume what when with workshop
http://acl.ldc.upenn.edu/P/P06/P06-2068.pdf	214	The Role of Information Retrieval in Answering Complex Questions	allan amigo answering bolivar buckley callan chua dang definitional detection document empirical evaluation evidence gaizauskas generic gonzalo greenwood hepple incomplete information level models novelty overview passage pattern peinado penas proceedings question references retrieval sentence sigir soft study synthesis task verdejo voorhees wade with workshop
http://acl.ldc.upenn.edu/P/P06/P06-2097.pdf	243	Unsupervised Topic Identification by Integrating Linguistic and Visual Information Based on Hidden Markov Models	advent alignment analysis annual applications associating association attention automatic babaguchi barbara barzilay baseball based broadcasted candace caption catching chang chih closed collaboration columbia computational conference content conversation cooking discourse drift eric extract foslerlussier from galley game generation gong grosz hamada hearst hidden hidehiko highlights hongyan huang ichiro icip ieee image intentions intermodal international into jing kathleen lillian linguistic linguistics march markov mckeown meeting michel models multi multimedia naacl naoko nitta noboru pages paragraph party passages peng probabilistic proceedings processing recognition references regina reiko related report sakai segmentation segmenting semantic shin shuichi sidner speech sports strategy structure subtopic summarization tanaka technical text textbook texttiling transcripts video winston with workshops yihong
http://acl.ldc.upenn.edu/P/P06/P06-1015.pdf	14	Espresso: Leveraging Generic Patterns for Automatically Harvesting Semantic Relations	aberdeen acquisition anlp automatic baltimore berland brown burdge bursten caraballo central charniak chemistry college corpora cover development downey edition elements etzioni finding from hall hierarchy hirschman hypernymlabeled information initiative john kozierok language large lemay mixed model ninth noun park parts prentice probabilistic proceedings processing redundancy references robinson science soderland sons systems text theory thomas very vilain washington wiley
http://acl.ldc.upenn.edu/P/P06/P06-2043.pdf	189	Improving English Subcategorization Acquisition with Diathesis Alternations as Heuristic Information	acquiring acquisition agustini alternations anna annual anoop association automatic barcelona cambridge chicago chinese classes clustering coling composition computational conference corpora czech diathesis dissertation distributions ellis english evaluation extraction filtering final folli frame frames from gabriel gamallo german germany group hall haoliang html http improving inducing information interest interface international keller korhon korhonen krymolowski language levin lexical lexicon linguistics lopes marx mccarthy meeting number oxford papers peters philadelphia please polysemic preferences press proceedings purely refer report resources runderstanding saarbrucken sabine sarkar science second selectional semantic semantically shulte siglex special subcat subcategorisation subcategorization sussex syntactic syntaxsemantics technical thesis tiejun tive trinity unit universitat university users using verb verbs walde with workshop xiwu yuval zeman zhao zvika
http://acl.ldc.upenn.edu/P/P06/P06-2006.pdf	152	Evaluating the Accuracy of an Unlexicalized Statistical Parser on the PARC DepBank	accuracy accurate american annotation annual arbor association bank based bikel boston briscoe budapest cambridge canaria carroll chapter clark cluk coling collins computational computer conf conference corpora corpus coverage crouch csli curran dalrymple deep dependency disambiguation dissertation driven efficient emnlp empirical english evaluation extraction formalization foundations general geneva gildea grammars grammatical gran granada head high hpsg hungary importance information international interpreted intricacies introduction inui iwpt japan kaplan keynes kiefer king klein krieger laboratory language lecture linc linguistically linguistics lrec malouf manning maryland maxwell meeting methods milton minipar miyao model models naacl natural north notes nunberg open oxford palmas parc parser parsing part pennsylvania performance pittsburgh precision press probabilistic proceedings processing proposal punctuation rasp references relations report resources riezler robust sampson sanfilippo sapporo schutze science sequence shallow sornlertlamvanich spain speech speed stanford statistical stochastic supertagging survey switzerland system systems tagging taipei taiwan tanaka technical techniques technologies text tokunaga tsujii university unlexicalized useful vancouver variation vasserman watson wide workshop
http://acl.ldc.upenn.edu/P/P06/P06-2093.pdf	239	Continuous Space Language Models for Statistical Machine Translation	approach bengio berger berghen bersini christian computational condor constrained della ducharme entropy extension frank hugues jauvin journal language learning linguistics machine maximum model natural neural parallel pascal pietra probabilistic processing references rejean research vanden vincent yoshua
http://acl.ldc.upenn.edu/P/P06/P06-1053.pdf	52	Integrating Syntactic Priming into an Incremental Probabilistic Parser, with an Application to Psycholinguistic Modeling	access accurate adaptation alan algorithm alterman american amit analysis anderson andreas annual architectures associates association based benedikt bock boston brants brian cache chance chapter charles christopher chuck church clifton closer cognitive computational computes conference contextfree coordinate coordinates coordination copy corner corpus corpuslinguistic creatures crocker daniel difficulty disambiguation down dubey earley editor editors efficient ellipsis empirical english erlbaum estimates evidence frank frazier from germany global habit hale hillsdale human ieee instance intelligence japan john johnson journal jurafsky kathryn keller kenneth kirsh klein kuhn language lawrence left lexical linguistic linguistics lynn machine manning mark matthew meeting methods model modeling mori munn natural noriegas north pages parallelism parser parsing patrick pattern persistence pittsburgh prefix priming probabilistic probabilities proceedings processing production psycholinguistic psychology rational recognition references renate research roark roland saarbrucken sapporo science sentence society speech spoken stolcke structures sturt syntactic syntax szmrecsanyi technology than that theory thorsten transanctions unlexicalized vancouver vanlehn widecoverage
http://acl.ldc.upenn.edu/P/P06/P06-1113.pdf	112	Question Answering with Lexical Chains Propagating Verb Arguments	aaai answering applications artificial baker based berkeley bernardo bonnie building canada challenge challenges charles chua class coling collin conference construction curran dagan dang dave entailment evaluation fifth fillmore framenet from gaithersburg glickman hang http innovative intelligence international james johan john karin keya kingsbury kipper kisuh language lexicon lowe lrec magnini main malvina march martha maryland montreal national network nissim november oren pages palmas palmer pascal paul press proceedings project propbank question recognising references renxu resources retrieval schuler seng seventeenth singapore spain tags task text textual trang trec treebank twelfth university using verb webber with workshop
http://acl.ldc.upenn.edu/P/P06/P06-2039.pdf	185	Parsing Aligned Parallel Corpus by Projecting Syntactic Relations from Annotated Source Corpus	abduction across after algorithm aligned aligning also although analyzing approach approaches assigned aswani based been beyond bombay bootstrapping bracketers building cabezas carnegie cases chatterjee clara computational computer concluding considered corpora corpus corresponding csli current daniel data datadriven david davy detail developing development editors engineering english esslli example experimental exploring focuses found gaizauskas general generalized goyal grace grammar have herein hindi hybrid identifying incompleteness india inducing induction information insufficiency italy khalil kolak language languages learning linguistics link links machine making mellon memorizing missing module more morphological morphology motivated multilingual naacl natural ngai niladri niraj noun october okan oliver oriented pages pair parallel parsefromspecial parser parsers parsing particular philip phrase present proc projected projecting projection proposed provides publications rebecca references remarks remko rens report resnik results robert robust rules scha scheme science sentence sentences september shailly should sima since sleator some specific stanford streiter structures study suffixes syntactic table taggers target technical temperley texts that these this towards translation trento university using version weinberg were which while with words work working workshop wrong yarowsky
http://acl.ldc.upenn.edu/P/P06/P06-1067.pdf	66	Distortion Models For Statistical Machine Translation	acknowledgment adam alberta alex algorithm alignment alignments also americas amta analysis andrew annual anoop apparatus approach april arabic association august automatic automatically bahl baltimore based basedword beam berger bleu boston brown canada center chapter christoph cocke coling college comparative complexity computational conf conference constraints context continuous contract copenhagen corpora cruz curin daniel darpa david decoder decoding dekai della denmark dragomir dumais editors edmonton empirical engineering english erhard estimation european evaluation features final franz franzjosef fraser frederick gale geneva gildea given hearst hermann hinrichs hopkins ieee improved improvements improving intelligence international jahr jain japan jelinek john johns joint josef july june katherine kehler kenji kevin khudanpur kishore knight koehn kumar lafferty lalit language large learned libin likelihood linguistics machine madrid main marcu mari marti maryland massachusetts mathematics maximum mccord meeting melamed mercer method methods michael model models monitored monotone naacl natural nist niyu noah number october onaizan ordering orientation ostendorf pages pami papers papineni parameter park partially patent pattern patterns paul peter pharaoh philadelphia philipp phrase pietra polynomial presentation proc proceedings processing program purdy radev recognition references reordering report rewrite richard robert roossin roth roukos salim sanjeev santa sapporo sarkar search september shankar shen short smith smorgasbord spawar speech states statistical stefan stephan stephen study submission summer supported susan switzerland syntax system this tides tillman tillmann time todd transactions translation under unigram united university using very vincent viren vogel ward washington weijing with word wordreplacement work workshop yamada yarowsky yaser zens zhen zubiaga
http://acl.ldc.upenn.edu/P/P06/P06-3009.pdf	280	Integrated Morphological and Syntactic Disambiguation for Modern Hebrew	adcock adler alon altman analyzer anthony approaches arabic automatique bank building carroll cassandra center charniak collins complement computational computer conclusion department design devising disambiguation disambiguator dudi eacl entropyinspired erel eugene evaluation fell gabai generative glenn gotoh habash haifa haim head hebrew helmut implementation instance institute israel isreal itai jeremy john katz khalil knowledge lack languages langues less lexicalised lexicalized linguistics littman lopar marking master maximum mccann meni metrics michael models modern morphological naacl nativ nizar object other owen pages parser parsers parsing part penn precludes presuppose probabilistic proceedings processing rambow references schmid science segal semitic sima size smoothing speech statistical stuttgart subject suited swoop taggers tagging technion techniques text texts than thesis they three tokenization traditional traitement tree treebank undotted university volume well winter yoad yoshihiko
http://acl.ldc.upenn.edu/P/P06/P06-4018.pdf	304	NLTK: The Natural Language Toolkit	adopted amund annotation applied astrid audience beasley bioinformatics biomedical bird comprehensive computational computer couples course courses data david documentation download dozens edition edward effective elizabeth essential extensive first framework google greid handson have hearst http including independent information interdisciplinary language learning lecture liddy linguistics links literature loper marti maximizes mccracken methodologies mining more nancy natural nltk notes pages please pointers potential practice proc processing provides python reference references rune sams science semantic sourceforge springer steigedal steven support teaching that theory tightly tonje toolkit tools tribulations triumphs tveit unique using volume which workshop
http://acl.ldc.upenn.edu/P/P06/P06-1003.pdf	2	Unsupervised Topic Modelling for Multi-Party Spoken Discourse	acoustics adam alex alfred allocation andrew annotations annual applications aspect banerjee barzilay based bayesian beeferman benefit berger blei browsing carolyn catching computer conference content david detect development dielmann dirichlet doug drift dynamic features generation hidden human icassp ieee information interaction international john jordan journal lafferty language latent learning level lillian machine main markov meeting michael model models moreno naacl necessity networks pages participants pedro playback probabilistic proceedings processing recording references regina renals research retrieval roles rose rudnicky satanjeev segmentation signal simple speech spoken state statistical steve structuring summarization system text topic using with
http://acl.ldc.upenn.edu/P/P06/P06-2026.pdf	172	Chinese-English Term Translation Mining Based on Semantic Prediction	annual based chen cheng corpora cross fang fifth finding from fung ijcnlp information language large mining nishino nonparallel proc queries references retrieval sigir teng terminology translating translation translations unknown very with workshop wvlc
http://acl.ldc.upenn.edu/P/P06/P06-1099.pdf	98	You Can't Beat Frequency (Unless You Use Linguistic Knowledge) ­ A Qualitative Evaluation of Association Measures for Collocation and Term Extraction	ananiadou annual applied approaches association august automatic balancing based beatrice bethesda better book bradford brigitte cambridge canada case chapter christian christopher clustering coling college collocation collocations combined combining complex computational conference daille dekang digital discovering eacl edition editors emnlp empirical enhancing european evaluation evert extracting extraction foundations france francisco frantzi frequency geneva goran hahn handbook hideki hinrich human identification implementation international jacquemin joachim john journal judith july june katerina kaufmann klavans krenn language lexical libraries library linguistics london lothar manning mass measures medical medicine meeting method methods mima modifiability montreal morg morgan multi national natural naught nenadic noncompositional october pages paradigmatic park philip phrases ppverb press proceedings processing qualitative quebec recognition references representations resnik retrieval sachs schutze similar sophia spotting springer statistical statistics stefan study switzerland symbolic syntagmatic system techniques technology term terminology terms than through toulouse unified value vancouver variation volume wermter word words workshop york
http://acl.ldc.upenn.edu/P/P06/P06-2044.pdf	190	Local constraints on sentence markers and focus in Somali	about accord activity advance advances african afrikanistische afroasiatic ahmed allows already also amsterdam anciennes andrzejewski annotated annual approach arabic arbeitspapiere aspects available basic been benjamin benjamins berkeley borovets boucher brown bulgaria buske called carried cascadilla case characteristics coling complement computational concise conclusions conference configurational configurationality congress constraints construction constructional constructions deriving developments discontinuous discourse dordrecht dowty editor editors elsevier encyclopedia engineering english fairly focus force form forms foth framework free from gebert gender general german gerunds given grimshaw hamburg have heads helmut hence illocutionary inclusion indicator inflectional ingo inquiry international introduction john keith killian kiss koskiennemi labahn language languages langues lecarme level linguistic linguistics linguistique literature logic logical louisa machinery malouf mansour many marseille meeting menzel mereu miller mirzaiean mitkov model montague morphographemics morphologies morphology morphosyntax most najmeh natural nicolo nominal notable number optimality order orientals other outlined oxford pages parse parsing particles persian peters phenomena philologie point polarity press proceedings processing production projection publishing puglielli quite ramsay ranlp recent recognition references reidel restrictif right robert robust role routine runs sadler saeed schaler schroder science second semantically semantics society sofia somali somerville speech studies svolacchia syntactic syntax text that theoretical theories this through topic treat treatment trees twenty university used vahid verbal verbs verlag wall weak weighted which with within wolfgang word work york
http://acl.ldc.upenn.edu/P/P06/P06-2072.pdf	218	Modeling Adjectives in Computational Relational Lexica	academic accidental address adjective adjectives adjectivo alonge already also amaro amherst analysis applications appropriately architecture argue argument ascribing association automatic base behav being bertagna besides between bosque building busa calzolari cambridge captured carlson case certain chaves chierchia cicling city classes cognition cognitive coherence complex complexos computacional computational computing conclusion conference conferences conhecimento constituency construction current dados database deficitary demonte descriptiva descriptive determines development dichotomy directions dissertation distinguish dordrecht eagles electronic encoding english enriching espa espasa eurowordnet event existencial existential explanation extraction feldweg fellbaum five focus fontenelle framework general german germanet ginet global gola gram grammar gross group hamp have implementation implemented information instituto intelligent interest interim internal international introduction island issue jeju journal kind kinds kluwer korea laboratory lenci lengua levi lexica lexical lexicalconceptual lexicography lexicon lexicons like line lingu linguistic linguistics lisboa lisbon lise lourosa lrec madrid marrafa martins massachusetts mcconnel meaning means mendes mexico millar miller milsark model modeliza modelling modifiers monachini mother motivation multilingual naacl network networks nevertheless nominal nominals observations ogonoski opposition orgs oxford palavra paper papers pecularities permanent peters portugu portuguese posici predica predicados predicates predication preliminary preserve press princeton proceedings processing properties property publishers pustejovsky recommendations reference references relational relations relevance report representation research resources revista roventini ruimy science seattle secund semantic semantics sentences simple sintagma small special stage stica stico strong structure structures syntactic syntax technical telic text that their theoretical this tica toward treatment university usos verbs villegas volume vossen which with wordnet wordnets workshop york zampoli zampolli
http://acl.ldc.upenn.edu/P/P06/P06-1051.pdf	50	Automatic learning of textual entailments with cross-pair similarities	abney academic acquisition advances alessandro alexander algorithms anal annie applied arbor automatic based bayer bernard bernardo bill bloothooft bmvc borovets boston boughorbel braz bulgaria burger cambridge canada carroll challenge charniak chierchia church circumscribed coling collins communications concepts conference conrath corley corpora corpus corpusbased courtney crouch dagan danilo darren database david defined discrete distance dolan dordrecht duffy eacl edit editor emnlp empirical engineering english entailment entropy equivalence eugene feature ferro fleuret france from generic gennaro george giampiccolo ginet girju glickman grammar grenoble guido haasdonk haim hearst henderson hyponyms idan ieee indefinite inference inspired intell interpretation introduction italy jason jiang joachims johan john karttunen katja kernel kernels kluwer koppel kouylekov language large lauri learning lexical lisa local logical mach magnini making mark markert marti maximum mcconnell meaning measuring mercer methods michael michelizzi michigan mihalcea milen miller mining minnen mitre model modeling morphological moschitti moshe naacl nantes natural nature nigel november object oren over parser parsing part partial pascal pattern patwardhan pearce pedersen perceptron practical press probabilistic proc proceedings processing publishers punyakanok rada ranking ranlp recognising recognition references relatedness richard rocling rodrigo roth roxana sally salvo sammons samuel scale seattle semantic semantics siddharth similarity southampton space speech springer statistical statistics steven structures submissions support svms szpektor tagging taiwan tapei tarel taxonomy text texts textual theory thorsten trans tree trento understanding vapnik variability vasin vector venice vladimir voted washington with wordnet workshop young zaenen
http://acl.ldc.upenn.edu/P/P06/P06-2024.pdf	170	Towards A Modular Data Model For Multi-Layer Annotated Corpora	bartsch based birmingham corpus eckart engineering form holtz july linguistics mechanical proceedings profiling references register teich texts
http://acl.ldc.upenn.edu/P/P06/P06-1106.pdf	105	Answer Extraction, Semantic Clustering, and Extractive Summarization for Clinical Question Answering	amia amigo annals answering aronson association being biomedical bulletin chambliss clinical cogdill conley covell dang databases dorsch effective emnlp empirical family first gonzalo groote information internal journal journals library manning mapping measuring medical medicine metamap metathesaurus moore needs office online overview patterns peinado penas practice program questions references resource responses scenario selection students study synthesis task text they uman umls verdejo workshop year
http://acl.ldc.upenn.edu/P/P06/P06-3005.pdf	276	Modeling Human Sentence Processing Data with a Statistical Parts-of-Speech Tagger	access acoustic acquisition algorithm ambiguity ambiguous amherst analysis animacy antecedents architectures asymptotically basic behavior bever bounds british burnard cambridge canada categories category charles clause clauses codes cognitive comprehending comprehension computational computing conference consortium conversational convolution corley corpus correcting crocker decoding differences dissertation down during empty error errors evensen evidence experimental exploring ferreira fixation foundations frazier frequency from guide harper henderson homes hypothesis ieee individual influence information inquiry international journal just king kuczaj language learning lexical linguistic linguistically linguistics macwhinney making manning massachusetts mcelree mechanisms memory model modeling modular montreal morris motivated movements national natural object optimal oxford paced parsing patterns pickering press probabilistic proceedings processes processing psychology rayner reading recognition references regan relative resolution roark role schriefers schutze seely self semantics sentence sentences service signal speech statistical stolcke strategies structurally subject syntactic syntax their theory transactions traxler trueswell university users verb verbal viterbi vonk wang word working
http://acl.ldc.upenn.edu/P/P06/P06-1118.pdf	117	Multilingual Legal Terminology on the Jibiki Platform: The LexALP Project	adaptation adapting alfa allows also amsterdam another antverpiensia antwerpen applications arntz article aspect august available bank base based benjamins between bilingual bistro browsing centralises choices coling collaboration collaborative communication company competence computer conclusion consortium construction contexte currently data database databases decisions definition description details developed development dictionaries dictionary dictionnaire different diffusion directed distribues doctorat does domain done editing editor editors entry environnements equivalence estfra estonian estonien european evolution evolve experimented felber files form fourier francais freely french gdef generic geneva gilles give gives grand great grenoble harmonisation helmut hence hoger hogeschool http information informatique initially instituut integrates interdisciplinary interface interlingual involvement isabella jibiki john joseph lafourcade languages legal leonhard lexalp lexical lexicale lexicographer lexicographes lexicologues liberty linguistic linguistica linguists loening lyding made makoto management mangeot manuel mathieu metadata microstructure monolingual moreover motivated multi multilingual multilingue nadia nagao natascia needs nouveau number oliver online organisation other pages papillon paris philadelphia pivot plan planning platform positive pour presented process programming project projet proved provided publishing ralli references regardless reiner relation require requires research resources resulting same science septembre serasset series services several shows some sonneveld spacial spatial specialite specific streiter structure structures structuring support sustainable switzerland taking team term terminological terminologie terminologists terminology terms that their theory therefore these this ties tighter tolken tools translation trivial unesco universite used useful users verena vertalers very voltmer volume voor when while will willing without
http://acl.ldc.upenn.edu/P/P06/P06-2046.pdf	192	Japanese Idiom Recognition: Drawing a Line between Literal and Idiomatic Meanings	accout accuracy acknowledgment adnominal aiming akio akira also ambiguity ambiguous among analyisis analysis annual applicable apply april asahara association atsukai baldwin believe belly between black bond brainwave broke cascaded cases chasen cheat chub chunking class collecting compound computational computer conclusion conference conll constraints content copestake could counting current daijiten decomposability decomposable dependency depends dictionary disambiguating disambiguation done douglas doushi doushikanyouku draw editors effective either equivalents evidence experiment exploits expressions fact fathom flickinger francis gakken gakushu geoffrey goes ground handle hara haruhiko have hayashi hearted hierarchy hirano hiromi hiroshi hiza http hyougen idiom idiomatic idioms ikeda ikehara implemented independent indicators information institute integrating intelligent international into ipsj ishida ishizaki issue ivan iwanami japan japanese journal jutsugo kagaku kaiseki kaisou kankei kanyouku kanyouteki kazuma keihanna kenji kenkyu kentaro kindaichi kitauchi kiyoko knee knowledge kokugo kosho kudo kuroi language learning lexical lexicon line linguistics literal long machine mackerela maintain manual masahiro masahito masayuki massachusetts matsuda matsumoto meanings meeting meiji miyaji miyazaki modification more morita morphological most multiword mwes nakaiwa nara natural neck negative nihongo nihongogaku nondecomposable nonpropositional notion nunberg ogura okeru ones ooyama operations other pages pain parsing perfect permitted phrase predicative priscilla problems proceedings processing proposed rate read recognition recognized recognizer references rejecting remained remains require requisite research restriction reveal revealed rohde saba satoru satoshi science selectional semantics sentences shirai shoin shoten showed shudo shun signl society some someone sousa soutou special speech still subclasses success syntactic system taikei taisuru takahashi takaoka taken taku tanabe tanaka tatsuo technique technology tedlab text tgrep than thank that their there things thinking third thomas those timothy toshifumi tougoteki transformability transformable transformations treatments uchiyama unable unambiguous usage used user using verb verbs version visible wasow well what when while with without workshop yamashita yasaburo yasuhito yokoo yomu yoshifumi yoshihiko yoshimura yoshitaka yoshiyuki youhou yuji yutaka
http://acl.ldc.upenn.edu/P/P06/P06-2115.pdf	261	From Prosodic Trees to Syntactic Trees	abney accents assisted basis between bfbs bible biblical british cambridge cantillation chanting database dresher edwin encoding foreign grosjean groves hebrew http institute jacobson jewish lampeter language lewiston lowery machine marks masoretes masoretic mellen morphology muenchen natural performance philadelphia phonology phrase press price proceedings prosodic publication published punctuation queenston references relation richter selkirk society sound speech structure syntax system team teamim their tiberian translation westminster workshop
http://acl.ldc.upenn.edu/P/P06/P06-4009.pdf	295	An Intermediate Representation for the Interpretation of Temporal Expressions	about alexander anniversary annotating annotation applications asian automatic bontcheva chen chinese cicling coling computational conference cunningham dale development douglas editor english environment events expression expressions february feng ferro framework gate gelbukh gerber graphical hacioglu hobbs information intelligent international interpretation irst jerry labeling language linguistics lncs local mani march marseglia maynard mazur meeting mitre negri normalization ontology pages proceedings processing reasoning recognition references report robust semantic semantics september society springer standard sundheim tablan technical technologies temporal tern text tides time tools transactions wilson workshop
http://acl.ldc.upenn.edu/P/P06/P06-1038.pdf	37	Efficient Unsupervised Discovery of Word Categories Using Symmetric Patterns and High Frequency Words	based berland brown charniak class comp corpora della desouza eugene finding jenifer language large linguistics matthew mercer models natural ngram parts peter pietra references robert very vincent
http://acl.ldc.upenn.edu/P/P06/P06-1050.pdf	49	Learning Event Durations from Event Descriptions	agreement algorithms analysis analyzing ando annotated annual applied approach artificial assessing assigning association automatic based boguraev brill british carletta choice classification clauses columbia commonsense computational computer conceptual conference congress constraints content context corpus data database decisions discourse dublin duda durations eacl edition estimation evaluation event events examples ferro fifth filatova fortemps francisco frank from fuzzy gaizauskas genoa gildea godo granularity grover half hanks hart hermjakob hitzeman hobbs hovy ieee ifsa ijcai imprecise intelligence international introduction ireland italy jobshop joint journal jurafsky kappa kaufmann kreinovich krippendorf labeling lancaster language lazo learning lexical lexicography line linguistics lingustics lrec machine magnitude mani meaning meeting memory methodology miller mining moens mooney morgan mulkar nafips natural nature news optimal orders parse part pattern possibilistic practical proceedings processing program programs publications pustejovsky quinlan radev reasoning references resources rich rieger robust roles rule sage saur scene scheduling semantic setzer simple spatial speech springer stanford statistic statistical structure sundheim systems tagger tasks techniques temporal text theory third timebank timemlcompliant timestamps tools transactions translation typical utterances vacouver vapnik verlag vila wiley wilson with witten wordnet workshop world york
http://acl.ldc.upenn.edu/P/P06/P06-1037.pdf	36	Guiding a Constraint Dependency Parser with Supertags	adjoining almost anlp applied approach bangalore based brants chen clark collins computational computerlinguistik conf constraint coverage curran daum deep dipper foth frameworks gram grammar hansen hendriks importance integration joshi kramp krenn language lezius linear linguistic linguistics meeting menzel models natural negraannotationsschema pages parsing part preis proc processing rambow references related report reranking saarlandes seattle shallow skut smith sozopol speech statistical supertagger supertagging tagger tech technical theories tiger tree treebank treebanks universitat using uszkoreit wide workshop
http://acl.ldc.upenn.edu/P/P06/P06-1033.pdf	32	Graph Transformations in Data-Driven Dependency Parsing	abeille academic accuracy accurate alena algorithm algorithms alon american annatual anne annotated annotation annual antal approximate aspects association barbora based basil bikel blackwell bohmova bosch brill building cambridge categorial chang chapter charles charniak chih christoph christopher chung classifier coling collins combinatory combining complexity computational conference constraints coordination corpora corpus corrective czech daelemans daniel data dependency diverse driven eacl editions editor efficient eisner elements english entropyinspired eric eugene european evaluation fernando first fourth gapping generative generator grammar grammars hajic hajicova hall hard head hladka hudson hyung igor improving independent informed international intricacies issues iwpt jarmila jason jens joakim johan johnson jong joon karolinum keith kenji klein klincksieck kluwer lance language lavie learning length leonardo lesmo level lexicalised library libsvm linear linguistic linguistics lombardo lucien machines maltparser manning mark maximum mcdonald meaning meeting memory michael model modeling models naacl natural nilsson nivre noah north novak online pages pajas panevova park parser parsers parsing pcfg pennsylvania pereira petr practice pragmatic prague press proceedings processing projective pseudoprojective publishers ramshaw references reidel representations resources richard ryan sagae scenario sentence sgall smith soft state statistical structurale support syntactically syntax syntaxe system technologies tesniere theories theory thesis three tillmann time tree treebank treebanks unit university unlexicalized using vaclav valency vector vidova vincenzo walter with word workshop york zabokrtsky zdenek zeman
http://acl.ldc.upenn.edu/P/P06/P06-1128.pdf	127	Semantic Retrieval for the Accurate Identification of Relational Concepts in Massive Textbases	available based blaschke boag chamberlin charniak coarse demo discriminative extraction fernandez fine florescu frame http ieee information intelligent johnson language line maxent medie module nbest parsing proc query references reranking robie simeon suiseki system systems tokyo tsujii valencia xquery
http://acl.ldc.upenn.edu/P/P06/P06-2028.pdf	174	Using Lexical Dependency and Ontological Knowledge to Improve a Detailed Syntactic and Semantic Tagger of English	adaptive algorithm analysis andrew annotated annual approach arivind arpa association barcelona based beatrice beyond bigram bilingual black bootstrapping brno broad building canada clustering coling collins collocation comprehensive computational computer conference copenhagen corpus coverage czeck daniel database demerdash dennis density dependencies developing disambiguation editor editors effective electronic empirical english entropy eubank evaluation ezra features fellbaum finch fourth francisco full garside general global grammars grammatical grinberg harabagiu hidden hideki hierarchical human iccpol ijcai international japan jean jersey john joshi journal july kashioka kaufmann kosseim kupiec lafferty lamjiri language large leech lexical linguistics link magerman marcinkiewicz marcus markov martha mary matwin maximum meeting merialdo methods michael mihalcea mitchell model modelling moldovan montreal morgan nancy natural pages pair palmer parser parsing part partof penn pittsburgh predictors press prithviraj probabilistic proc proceedings processing producing proving publishers rada ramakrishnan ratnaparkhi references reinventing report republic robust rosenfeld saia sanda santorini scale semantic sense senseval simple skeleton sleator soft somerset spain speech stan state statistical suarez syntactic system systems szpakowicz tagger tagging tagsets technical technology text third thirty translation treebank trigger ushioda using with word wordnet words workshop xiaobin yarowsky
http://acl.ldc.upenn.edu/P/P06/P06-1077.pdf	76	Tree-to-String Alignment Template for Statistical Machine Translation	alex alignment alshawi americas andreas annual approach arul ashish association automatic bangalore based beam better bilingual bleu brown center chen cherry chiang chinese chris classification colin collections computational computing conference considerations corpora daniel david decoder dekai della dependency deyi ding discriminative douglas eamt emnlp empirical entropy error estimation european evaluation extensible finite fourth franz galley goodman graehl grammars harvard have head hermann hierarchical hiyan hopkins ijcnlp improved improvement information informed insert international interpreting inversion joint jonathan joshua kenji kevin kishore knight knowledge koehn language learning linguistics loglinear lrec machine marcu mark martha mathematics maximum meeting melamed menezes mercer method methods michel minimum model modeling models much mutual naacl natural need nist pages palmer papineni parallel parameter parsing penn peter pharaoh philipp phrasal phrase phrasebased pietra probabilistic probability proceedings processing qian quirk rate references report research resources robert roukos rule salim scores search semantic shona shouxun shuanglong sixth smoothing spoken srilm srinivas stanley state statistical stephan stephen stochastic stolcke study synchronous syntactically syntaxbased system technical techniques technology template tenth todd toolkit training transducers transduction translation tree treebank treelet trnaslation university using venugopal vincent vogel volume waibel ward weijing what william with wong word xiong yamada yang ying yuan yueliang zhang
http://acl.ldc.upenn.edu/P/P06/P06-1057.pdf	56	Direct Word Sense Matching for Lexical Substitution	advances automatic bernardo burges canada challenge challenges chapter chugur cigarran clustering cone conference cream dagan database dekang dictionaries disambiguation editors electronic entailment fellbaum from glickman gonzalo improve indexing joachims kernel large learning lesk lexical machine magnini making methods montreal oren pages pascal pine practical press proceedings readable recognising references retrieval scale scholkopf sense sigdoc similar smola support synsets tell text textual toronto using vector verdejo with wordnet words workshop
http://acl.ldc.upenn.edu/P/P06/P06-1039.pdf	38	Bayesian Query-Focused Summarization	allocation andrew applications bayesian blei bruce buckley callan chengxiang chris conference croft daniel database daume david developments dirichlet document evaluation expert extrinsic feedback gerard harding information inquery international intrinsic jamie january jmlr john jordan journal lafferty language latent learning machine marcu measures michael minimization models multi optimization proceedings query references relevance research retrieval risk salton sigir stephen suggestion summarization system systems understanding weights workshop zhai
http://acl.ldc.upenn.edu/P/P06/P06-1065.pdf	64	Improved Discriminative Bilingual Word Alignment	abraham advances alexander algorithms aligner alignment alignments annual arabicenglish arbor ayan bayes bonnie british brown building cherry christof colin collins columbia combining computational conference daniel dekang della discriminative dorr empirical english entropy estimation experiments fazil fraser graepel herbrich hidden human improve information ittycheriah japan language large linguistics machine machines marcu markov mathematics maximum meeting mercer methods michael michigan model models monz natural necip networks neural neuralign parallel parameter participation pennsylvania perceptron peter philadelphia pietra point probability proceedings processing ralf references robert romanian roukos salim sapporo scale statistical stephen systems task technology texts theory thore training translation using vancouver vincent with word workshop
http://acl.ldc.upenn.edu/P/P06/P06-2091.pdf	237	Translating HPSG-style Outputs of a Robust Parser into Typed Dynamic Logic	about achieved achieving acquiring adequacy also although ambiguities anaphora appear approximately arpa athens baldwin beast beauty beavers bekki bender between beyond broad carl chicago clark cleaner coling compact compositional computational conclusion construction contemporary copestake corpus coverage criterion curran daisuke dealing deduction descriptive design development difference discourse dissertation doctoral driven dynamic each emily empirical english environment essential evidence expressive extracting flickinger from geneva given govern grammar grammars groenendijk gruyter have head high hockenmaier hpsg human ichi ijcnlp implementation incompatible inter into intrasentential intrinsically introduction investigated involving ivan james jeroen johan john journal julia language laws levels lexicalized linguistic linguistics lnai logic london lrec manuscript marcus mark martin method might minimal mitch miyao mouton natural ninomiya notion oepen only open oriented outputs over palmas paper parser penetrates penn perspectives philosophy phrase plurality pollard possibility power practical precision predicate predicateargument present press presupposition previously princeton principle proc proceedings processing proposed proved quantification real reconsidered recursion references relationship representation representations resulting revised reyle robust robustness running semantic semantics sentential serves showed source springerverlag steedman stephan stephen stokhof structure studies study style sufficient superficial syntax takashi taught technolog text that then theoretical theories theory thought timothy tokyo trade translate transparency treebank tsujii typed underspecification university usefulness using validity what which widecoverage with workshop yusuke
http://acl.ldc.upenn.edu/P/P06/P06-1012.pdf	11	Estimating Class Priors in Domain Adaptation for Word Sense Disambiguation	aaai adam agirre algorithms annals automatically ayer based bayesian beyond bias brunk carroll chan chklovski classifier conditions david dependence description diana disambiguation distribution domain domingos edward emnlp empirical eneko english escudero estimation evaluation ewing examples finding function gerard german hwee icml ijcai importance incomplete independence information john julie keok kilgarriff knowledge koeling learning lexical lluis marquez martinez mathematical mccarthy michael mihalcea miriam optimality parallel pazzani pedro predominant proc rada references reid retrieved rigau sample sampling scaling seng sense senses senseval silverman simple sources statistics study supervised systems task text texts timothy unsupervised untagged weeds with word yoong
http://acl.ldc.upenn.edu/P/P06/P06-2052.pdf	198	Efficient sentence retrieval based on syntactic structure	algorithms alick already amsterdam analogy analysis annotate annotated annotation annotationg annotators approximating artifical asahara asian back banerji based between bombs bombshell california came classroom cognitive coling collins company compared conclusion conference convolution corpora corpus cruz csnlp dependency derived developing dublin duffy ebonsai edition editors elithorn enables english estimating evaluation example experiments exploded extended fast faster figure following framework from generation good group hashimoto have head high human information instance instances integrating integration intelligence interest international inui iwabuchi iwanami iwanamishoten japan japanese jiten jones journal kernel kernelbased kernels koike kokugo language looking lrec material matsumoto mclean measures mechanical methods mizutani models mslr multilingual nagao named natural neuron nips nishio noro obtain outputs overlapping pages parser parsing part performance piece principle problems proceedings processing proposed ranan ranke references report resources respect results retrieval retrieve retrieved retrieving room santa science sentence sentences share shirai similar similarity single society somers special speech statistics structural structure structures subpath supports syntactic syntactically tagger takahashi tanaka teaching technical techniques than them times tokunaga tool tools translation tree ucsc ueki university useful which with workshop xplode yoshida young
http://acl.ldc.upenn.edu/P/P06/P06-1094.pdf	93	Proximity in Context: an empirically grounded computational model of proximity for processing topological spatial expressions	baldridge budapest categorial combinatory coupling dependency eacl grammar hungary hybrid kruijff logic modal multi pennsylvania philadelphia proceedings references semantics
http://acl.ldc.upenn.edu/P/P06/P06-1056.pdf	55	Semi-Supervised Learning of Partial Cognates using Bilingual Bootstrapping	acquisition algorithm applicability association bilingual carroll cognates computational context corpora diab disambiguation gass hearst homograph identifying issue jacques journal language large lexicon linguistics lists local machine marti meeting method mona noun parallel philadelphia philip proceedings quantitative references research resnik second sense special studies susane tagging text translation unsupervised using word
http://acl.ldc.upenn.edu/P/P06/P06-2025.pdf	171	A Modified Joint Source-Channel Model for Transliteration	abdul abstract algorithms annual approaches arabic arbabi bilingual cheng cikm computational conference crego cross development elizabeth english entities entity extended fischthal gispert human information international jaleel journal knight knowledge language languages larkey leah machine management mansur marino meeting monolingual name named names nasreen ngram onaizan orleans proceedings references reordered research resources retrieval scott search semitic statistical technology text translating translation transliteration tuple twelfth unfolding using vincent workshop
http://acl.ldc.upenn.edu/P/P06/P06-1070.pdf	69	Exploiting Comparable Corpora and Bilingual Dictionaries for Cross-Language Text Categorization	acquiring aligned american analysis arbor august barcelona bilingual building burch callison categorization comparable computer conference conjunction corpora cross crosslingual dagan deerwester dejean digital disambiguation domain domains dumais ecdl european exploitation extraction from furnas gaussier geometric gliozzo goutte harshman indexing information journal july june koster landauer language latent lexical lexicon libraries machine matveeva michigan models multilingual osborne parallel proc proceedings references renders science semantic sentence society spain speech statistical strapparava supervised talbot text texts translation trondheim university unsupervised using view villegas with word workshop
http://acl.ldc.upenn.edu/P/P06/P06-1092.pdf	91	Phoneme-to-Text Transcription System with an Infinite Vocabulary	adam advances alfred algorithm algorithms almost analyzer anne approach associates automatic backward based bazzi bernard best boundary brown center channel chapter church cjkv cocke coling complexity computational computer conclusion conversion corpus correction daisuke dekker della derouault elseveir emnlp estimation extraction finding forward frederick from gale generalized generative glass gram grefenstette gregory handbook icslp ieee infinite informa information introduction ipsj issam issue james japanese jelinek john kana kanji kenneth kernighan kilgarriff lafferty language languages lexical linguistics lunde machine makoto marie mark masaaki masatoshi mercer merialdo model modeling mori morphological nagao nagata natural noisy obvious organized osamu outof pages pami paper patterns paul peter phoneme pietra principles probability proc processing program proposed publishers recognition references reilly report research robert robust roossin roukos salim science search self shinsuke signal special speech spelling statistical stephen stochastic strings system takuma technical text texts theoretical this transactions transcription translation tsuchiya using vincent vocabulary volume watson william with without word words yamaji
http://acl.ldc.upenn.edu/P/P06/P06-1082.pdf	81	Word Alignment in English-Hindi Parallel Corpus Using Recency-Vector Approach: Some Studies	about academic across activities actual advertisement ahrenberg aided algorithm algorithms aligning alignment already alternative americas ananthakrishnan annual approach approaches asian association athens available barrier based been beneficial bilingual book brown budapest certain chapter chinese choi church clues columbia columbus combining comparison computational conclusion conference consequence considered considers constraint corpora corpus correct correspondences could crossing dagan darpa della dependent developed developing different distinct durgesh dynamic eacl engish english estimation europe european evaluation expanding experimental experiments extraction feature figure first focuses fourth freely frequency fung further gale given good greece groups hand have hein highly hindi hong however huang hungary identifying improve improved increase industrial initial international involving japanese jorg journal kaufmann kong korean language languages large level lexicon lightweight like linguistic linguistics lrec machine manifold maryland matching mathematics mckeown measures meeting mercer merkel morgan natural noisy ohio order other pages pair pairs paper parallel parameter partnerships paucity performance performances perspectives pietra proc produce proposed prove provide publishers ramanathan range recency references resource resources results rich robust root sagvall schemes score segment sentence size somers south speech statistical stemmer story storybook such suggests suitable synergies system systems table taken task techniques technology text texts that their these they this thousand three tiedemann time tools total translation used variations various vector very vocabulary volume warping where with word words work workshop
http://acl.ldc.upenn.edu/P/P06/P06-2058.pdf	204	Obfuscating Document Stylometry to Preserve Author Anonymity	acapulco acknowledgements addisonwesley addition advances alberta american analysis anonymous applications approach approaches artificial attribution authorship banff bayesian bosch brockett burges cambridge canada categorization chickering chris class classification colorado computational computers computing conference denver directions disputed document drafts earlier european evolution exploiting fast features federalist first forsyth guarantee heckerman holmes humanities hyperplanes icml idiosyncrasies ijcai inference information intelligence international joachims kernel koppel learning linguistic literary lkopf local machine machines many massachusetts mathematical meek methods mexico minimal monthly mosteller network networks neural notably optimization order papers platt press privacy problem proceedings provided providence pseudonymity reading really references regard relevant report reviewers revisited rohatgi schler scholarship security separating sequential singh smith smola structure style stylistic stylometry support svms symposium synthesis technical text thanks thirteenth this toolkit training tweedie twenty uncertainty useful usenix using vector verification wallace winmine with workshop
http://acl.ldc.upenn.edu/P/P06/P06-2012.pdf	158	Unsupervised Relation Disambiguation Using Spectral Clustering	acmdl advanced advances agency agichtein algorithm algorithms among analysis andrew annual aone applied april association automatic barcelona brin brown charniak cikm classification clustering clusters collections combining computational computer conference connecticut corpora culotta cuts database defense department dependency determination digital ding discovering eigenvectors emnlp empirical entities entropy exploring extending extract extracting extraction features foundations from good gravano grishman guodong hasegawa iccv ieee image information inspired intelligence international jian jordan july kambhatla kannan kaufmann kernel kernels knowledge laidler language large lawrence learning lexical libraries linguistics machine malik management maximum means meeting message methods miller models morgan mystic named natural neural nips normalized novel number pages parser parsing pattern patterns philadelphia plain proc proceeding proceedings processing projects publishers ralph ramshaw references relation relations relaxation report research richardella sanguinetti satoshi science seattle segmentation sekine semantic sept sergey signal simon sixth snowball soresen spain spectral statistical supervised syntactic systems takaaki technical technology text transactions tree understanding unifying university using various vempala vetta view washington weakly webdb weischedel weiss wide with workshop world yair zelenko zhang zhou
http://acl.ldc.upenn.edu/P/P06/P06-4002.pdf	288	Is It Correct? - Towards Web-Based Evaluation of Automatic Natural Language Phrase Generation	aaai accessibility alessio alicebot answering antonio applications arendse artificial based bernth between billion billmann cabral california claw cody common commun communication comput computer conference controlled corpus cynthia daniel dave ddivers dialogue diversity doug easyenglish elizaa etzioni foundation from fully generation giles gulli http indexable information intelligence international inui john joseph junpei kahlert koiso kotani kwok language lawrence lenat linguist machine matuszek michael more nakamura national natural nature noah nobuo oren pages parallel philip populating preprocessing proceedings program purvesh question references resnik richard robert scaling schneider searching second sense shah signorini site smith south spoken spring steve study symposium syst system takuya than trans twentieth university wallace weizenbaum weld wide witbrock workshop world written wwwrcf yoshiyuki
http://acl.ldc.upenn.edu/P/P06/P06-1032.pdf	31	Correcting ESL Errors Using Phrasal SMT Techniques	bond brown coling computational countability della francis ikehara japanese kentaro linguistics machine mathematics mercer number ogura peter pietra references robert satoru statistical stephen toenglish translation vincent
http://acl.ldc.upenn.edu/P/P06/P06-1103.pdf	102	Weakly Supervised Named Entity Transliteration and Discovery from Multilingual Comparable Corpora	aaai abduljaleel academic advances alexandre alignment ambiguities ambiguous american annual approach approaches arabic arfken articles artificial association attribute avrim bilingual blum boolean brain chapter cikm classification coling collins combining comparable computational conference contextual corpora cross cucerzan data databases david discovery discriminative efficient emnlp empirical english entity eunok european evidence extended extragradient framework frank from functions generative george gerard graehl hetland hong identification ijcai independent infinite information intelligence international introduction joint jonathan jordan julien jung kevin klementiev knight korean lacoste language larkey leah learning linguistics machine magnus markov mathematical mcgill mcgrawhill meeting method methods michael mining model models modern moore morie morphological multilingual naacl named names nasreen national natural neural news nips north organization paek pages paul perceptron physicists prediction press probabilistic proc proceedings processing psychological recent recognition references resolve retrieval review robert rosenblatt roth salton satoshi scientific sekine sequences series shinyama silviu similar simon singer space statistical storage structured sung sunglim survey systems taskar time tracing transliteration unified unsupervised using window word world yarowsky yoram york young yusuke
http://acl.ldc.upenn.edu/P/P06/P06-1018.pdf	17	Polarized Unification Grammars	abstraction ajdukiewicz alternative applications approach baltin based blackwell bonfante bresnan burroni chicago coling computer conceptions conference constituent constraint cruces dependencies descriptions dimensional disambiguation distance duchier dymetman equational formalisms functional geometry girard grammar grammatical guillaume higher interna kahane kaplan konnexit kroch language lareau lexical linear logic long mathematics meaning meeting methods modularity moscow natural nlulp orlando parsing perrier philosophica phrase polarization press problems proceedings programming references remarks sciences second some structure studia switzerland syntaktische syntax text thater theoretical theory tional tree uncertainty understanding unification univ with word zaenen
http://acl.ldc.upenn.edu/P/P06/P06-2032.pdf	178	Coreference handling in XMG	abeille adjoining adjoints amsterdam application arbres aspects based becker candito coling colloquium computational constraint copenhagen crabbe csli cslp descriptions duchier editors formal fortement gardent gram grammars hierarchical informatique iwcs kopenhagen lexicalisees linguistic logic ltags maire maires metagrammar metarules nancy patterns predicate principle proceedings publications rambow redux references representation stanford thesis tilburg treatment tree universite unplugged
http://acl.ldc.upenn.edu/P/P06/P06-2065.pdf	211	Unsupervised Analysis for Decipherment Problems	algorithm approach artificial automatic basil behaviour blackwell blevins breaking brown cambridge categories chadwick chater code computational data decipherment della dempster editor estimation finch from goldsmith handbook hudson hybrid incomplete intelligence journal laird learning likelihood linear linguistic linguistics london machine mathematics maximum maya mercer parameter phonological pietra press quarterly references royal rubin simulated society statistical syllable thames theory translation university york
http://acl.ldc.upenn.edu/P/P06/P06-1124.pdf	123	A Hierarchical Bayesian Language Model based on Pitman-Yor Processes	advances american analysis annual association bayesian bengio between breaking carlin chapman chen chinese computational computer conference data dirichlet ducharme empirical entropy estimating exponential gelman generators ghahramani gibbs goldwater goodman griffiths group hall harvard information interpolating ishwaran james jauvin johnson jordan journal language learning linguistics london machine maximum meeting methods microsoft model modeling models neural nips nonparametric power presentation priors probabilistic proceedings processes processing progress references report research restaurant rubin sampling science smoothing statistical stern stick study systems technical techniques that tokens tutorial types university vincent volume
http://acl.ldc.upenn.edu/P/P06/P06-1062.pdf	61	A DOM Tree Alignment Model for Mining Parallel Data from the Web	accurate acquisition across adaptive adjoining algorithm align aligning alignment alshawi american americas analysis annual articles association australasian automatic bangalore bannard based bilingual bits bottleneck brown build burch callison chan chapter char character chau chen chinese church collection collections comparable complementarities computational computer conclusion conference context corpora corpus criterias cross data dependency development disambiguation discovering discussion douglas durand dynamic empirical english estimation european exploiting extraction faced fast feature features fields final finite first fraser free from fully fung gale generation grammars groups hajic head human ieee improved inducing information insideoutside intellig intelligence international internationalization inversion isabelle isahara issues japanese knight knowledge lafferty language lari learning level lexical liberman linguistics machine marcu matching mathematics mckeown measures meeting mercer method mined mining model models moore munteanu naacl natural news noisy north over overcome pair parallel parameter paraphrasing parsing pattern performance phil pietra practical proceedings program promising quality random references reliable report research resnik results retrieval roscheisen schabes search second security sense sentence sentences shieber should sigir smith software speech state statistical statistically stochastic studied study summit synchronous syntax system table technology text texts three time transactions transducers transduction translation tree using utiyama vines vogel wang warping website wide with word workshop world yamada young zhang zhao
http://acl.ldc.upenn.edu/P/P06/P06-1146.pdf	145	Optimal Constituent Alignment with Edge Covers for Semantic Projection	acknowledge acknowledgments across acta algorithm algorithms aligned aligning alignment alignments allow allows although analyses analysis annotation approach approaches arbor aspect assignment assumption augmenting authors automatic automatically background backoff barcelona based between biframenet bilingual bille bipartite blinker boas bootstrapping boston calibrating carreras chen clause coling collins comparison computation computational computer computing conclusions conll constituent constituents construction corpora corpus correspondence cover covers cross databases defined deletions dense dependencies determined diego distance divergences driven dubey eacl edge edit edmonton efficient efficiently eiter emnlp english enough equivalence essential europarl evaluation explicitly fails falls features fibonacci fillmore filtering formalisation formalisms frame framenet frames fredman from fung furthermore future generative geneva german gildea good grant graph have heaps heuristic however improved improving inducing induction informatica information insertions interlingual international investigate ircs japan johnson jonker journal jurafsky koehn kolak kucerov labeling labelling languages lapata learned lexical lexicalised lexicalization lexicography linear lingual linguistic linguistics loosely machine madrid mannila manual marcu mccord measures mechanisms melamed method minimal minimum model modelling models multilingual naacl network ngai noise novel obtaining open optimization palmer paper parallel parsing path patterns pennsylvania performance petruck philadelphia phrase phuket point problems proceedings project projection proposed provides question references related report representations resnik resources restructuring rests results rewrite robust role roles rquez sapporo science search semantic semantics sentences sets shared short shortest shown similar similarity smoothing solely spain sparse statistical structure structures substitutions such suffix summit superior support survey switzerland syntactic system systematic tarjan task technical techniques text thailand that their theoretical therefore this three tools translation translational tree treebased university useful uses using vancouver various volgenant weight weinberg what when whether which will with work yarowsky yields
http://acl.ldc.upenn.edu/P/P06/P06-1100.pdf	99	Ontologizing Semantic Relations	agirre ansa applications arbor artificial basili cafarella church coling computers concepts conceptual copenhagen corley corpus customizations danmark database density disambiguating disambiguation downey driven ecai electronic empirical enriching entailment entity equivalence etzioni event experimental extensions extraction fellbaum from gale hovy humanities information intelligence large learning lexical machine martinez measuring method mihalcea modelling naacl named other pazienza pittsburgh popescu press proceedings recognition references resources rigau rules semantic sense senses shaked signatures similarity soderland study texts topic unsupervised using vindigni weld with word wordnet workshop yarowsky yates
http://acl.ldc.upenn.edu/P/P06/P06-1139.pdf	138	Stochastic Language Generation Using WIDL-expressions and its Application in Machine Translation and Summarization	adjoining algorithms amalgam amta approach artificial bangalore bonnie charles clifford cmejrek columbia computer context cormen corston cucs daniel david decoding department ding dorr eisner elhadad eric fast fifth final gamon generation germann ghmt gildea grammars habash hajic headline hedge hill hltnaacl hopkins inlg intelligence international introduction jahr johns kenji kevin knight language large learned leiserson machine manual marcu matador mcgraw michael mike model module moore natural nizar oliver optimal overview owen pages parse parton penn press proceedings radev rambow references report richard ringger rivest robert ronald scale schwartz science simon spanishenglish srinivas stein summarization summer system technical text thomas translation tree trim trimmer ulrich university user using version workshop yamada zajic
http://acl.ldc.upenn.edu/P/P06/P06-4003.pdf	289	LeXFlow: a System for Cross-fertilization of Computational Lexicons	according adriana alessandro alonge andrea antonietta antonio apache authentication automatic based bernardo bertagna best between browser building calzolari capabilities capable centric challenge christian cignoni client cocoon communicate communication component computational computer concept connotes consists container cristina database databases defined deliverable displayed displaying document documents each edit editoriale editors electronic elisabetta encrypted entire entry events finally fiorentino form framework francesca functioning girardi gola implemented interface internazionale isle istituto italian italwordnet language large laura lenci lexical lexicon lexicons linguistics linking magnini marchetti marinelli marisa maurizio mile minutoli modified module monachini monica multilingual mysql needs nicoletta nilda pages pipeline pipelines pisa poland poligrafico power poznan practice proceedings process publishing receive references rendered request rita roma rossi roventini ruimy salvatore science semantic send sergio series server servlet side simple specified standards taking technologies template tesconi that then through tomcat towards treatment type ulivieri used user uses using when which will wise with workflow xflow xforms york zampolli
http://acl.ldc.upenn.edu/P/P06/P06-1004.pdf	3	Minimum Cut Model for Spoken Lecture Segmentation	accuracy across advanced advances aiming algorithm algorithms align aligning analysis anisotropic annotation application applications approaches automatic automatically beeferman been believe berger between boost buckley cettolo char character choi church cohesion combining components computational computer conclusions content conversation corpus could critique current cuts data dependencies depending describe determine diagnostic dialogue diffusion discourse dissimilarity distribution does domain dynamic each eacl emnlp employs english errors evaluation existing explore explored expository federico fosler fragkou framework further galley generate generating generation glass global goal granularity graphpartitioning gruenstein halliday hasan hastings hearst icassp ieee image impact implementation improve improvement improves independent information intelligence interaction investigate isahara issue jing kehagias lafferty language latent learning lectures leeuwis level levels lexical linear linguistics london long longman lussier machine malik management mckeown measuring meeting methods metric model modeled modeling models moore multi naacl needs niekrasz normalized optimize paper paragraph parallel party past pattern pennsylvania performance petridis pevzner plan porter presence probabilistic proceedings processing produce program programming purver range recognition references resulting retrieval reynar robust salton science segment segmentation segmentations segmentbased segments semantic showed sigdial sigir similarity simultaneously smoothing speech statistical strategies stripping structure studied succinctly such suffix swets system systems tables task term text texts that thesis this titles tools topic total transactions transcription ultimate university using utiyama various weighting whole wiemer will with within workshop
http://acl.ldc.upenn.edu/P/P06/P06-2050.pdf	196	When Conset meets Synset: A Preliminary Survey of an Ontological Lexical Resource based on Chinese Characters	academia acknowledgements additional against agreed aitchison albert also analyses annual anonymous application approach augment authors avenues barabasi base based been blackwell bong both cambridge canonical case cbflabs character characteristics characters china chinese chou chris claim cognitive collective coming comments common communications complexities computational concepts conceptual conceptualization conclusion conference consider constitutes construction constructions constructive contemporary conventionalized data decisions deep development different dynamics emergence enriched evaluating existing feng field first found from future general germany goal groundwork growth guarino hantology hanzidriven hanzinet hovy hsieh http huang iccs ideographs ijinlp immaturity important incorporated incremental inducing information insights instead institute integrated integration interaction interfacing international interpretational into introduction island itself jean jijeu kassel kindly knowledge korea language largescale laszlo leads level levels lexical lexicon like linguistic linguistics manifestation many mapped mathematical meaning mental methodologies might mind model modern morpheme morphology mugnier natural nature needed network networks nicola normally noted ontoclean ontolex ontological ontologies ontology order other packard perspective press problems proceedings processing propose provide publishing radicals random referees references reka reliable representation represented require research resource resources scaling scenario science scientific script seated seek semantic semantics several sharing shiwen should sighan sinica small south statistical steyvers strogatz structure structures stumme subjectivity sufficient support survey syntactical system systems tasks tenenbaum thank thanks that their theories therefore thesis this though three topic tradition treatments under understanding understood undervalued unfortunately unique university upon using various views watts ways well welty which will with word wordnet words workshop world would writing xiamen xuefeng
http://acl.ldc.upenn.edu/P/P06/P06-1137.pdf	136	Highly constrained unification grammars	applicability approximation attribute barton based berwick california cambridge carpenter chapter christian cognition complexity computational conference constraint corner csli daniel editors edward efrat eric feature feinstein finite francez gazdar gerald grammar grammars haifa indexed information international investigation jaeger johnson journal language languages lecture left line linguistic linguistics logic mark master models natural nissim notes pages parsability parsing perception press proceedings references reidel reyle ristad robert rohrer shuly stanford state structures sven theories theory thesis transforms typed unification university using value volume wintner
http://acl.ldc.upenn.edu/P/P06/P06-2092.pdf	238	AT L A S	able across adding align aligning alignment amazing americas amta annotation annotator annual appear approach association automatically based berkeley bettina bilingual birmingham bootstrapping brown budapest california cambridge canada chapter cherry china christoph christopher chunker church clues cognates cognitive colin coling collocations columbia combining comparable computational conference copenhagen corpora corpus create croco cross currently decision dekang della denmark dictionary dynamic eacl emnlp empirical england english estimation europarl european evaluation explicitation exploiting expressions extensions feature fifth first foster foundations fourth frank franz french fung further gale geneva genoa german grammar groups guide hansen hatzivassiloglou helmut hermann hiemstra hinrich hong hungary ilhan improve improved induction institute international interpreted investigateon isabelle issues italy japan jennifer jonas jorg josef kathleen kenneth knowledge koehn kong kristina kuhn kyoto language languages level lexicons linc linguistic linguistically linguistics link linkoping lisbon london lrec machine magnus manchester manning manuscript march martin maryland massachusetts master matching mathematics mckeown meeting melamed mercer merkel methodological methods michel model models monolingual montreal natural neumann noisy nominal nordic norway osnabruck pages pair pairings parallel parameter part pascale patrick pennsylvania peter philadelphia philipp pietra pilot portugal possible press probabilistic probability proceedings processing program project rare references report reprinted research resources robert roscheisen samuelsson sapporo schirra schmid schrader schutze science sentence sentences silvia simard sitat smadja speech statistical stella step stephan stephen study style such summit supported swedish switzerland system tagging technical texts texttranslation that theoretical thesis tiedemann tillmann time tolga toutanova translating translation treebanks trees trondheim tschorn twente univer universiteit university unpublished using utility vasileios vincent vogel volk warping will william with word work workshop yvonne
http://acl.ldc.upenn.edu/P/P06/P06-1130.pdf	129	Robust PCFG-Based Generation using Automatically Acquired LFG Approximations	aberdeen abney acquired acquisition adams ambiguous andy anja aoife approximations asia attribute automatically baayen bangalore barcelona based belz bodomo burke cahill chan chen chinese coling compared computation computational conference corpora dependency distance donovan emnlp enlg estimating european evaluated exploiting forms frequency functional genabith generation germany grammar grammars harald hierarchical impact information japan john josef language lexical linguistics long methods michael model morphologically natural olivia owen pacific pages pcfg priors probabilistic proceedings quality quantity rambow references resolution richard rowena ruth saarbrcken scotland spain sproat srinivas statistical stephen stochastic three tokyo treebank value widecoverage workshop
http://acl.ldc.upenn.edu/P/P06/P06-2048.pdf	194	Exploring the Potential of Intractable Parsers	aaai accurate advances adwait although another approaches assigned attempt august available based basic been began bfgs bigram black both brill carefully case cgbfgs charniak chosen christopher collins computational considered constraints context daume david decision dependencies develop directions discussion docs driven dynamic effective emnlp employed entropy eric eugene evidence experiment explored extend ezra favorable feature features first flexible framework fred free from general given gives grammar grammars hdaume head history http immediate implementation indeed information inspired instance intractable jelinek john johnson klein lafferty language leaves left lexical linear linguistic linguistics logistic magerman manning mark maximum megam mercer methods michael mind model models more most multiple naacl natural next notes observed only order paper parameter parser parsing part pcfg pennsylvania perform places practice presented probabilistic proc programming project provide quality question ratnaparkhi references regression representations research richer right robert roukos rule running salim selected sense simple simplicity some sought specific speech statistical statistics step successful tagging terms that there thesis this time timization towards tree unadorned university unlexicalized used using variables versatile well were which with word work worst would
http://acl.ldc.upenn.edu/P/P06/P06-1080.pdf	79	Self-Organizing Ò-gram Model for Automatic Word Spacing	algorithm annotation annual association automatic bengio bidirectional block charniak chen computational data detecting dickinson discontinuous dortmund ducharme empirical eojeol errors estimation from goodman hangul interpolated jauvin jelinek joachims journal kang kiss korean language large learning linguistics machine making markov meeting mercer meurers model modeling neural parameters pattern practical practice press probabilistic proceedings recognition references research scale sentences smoothing source spacing sparse statistical structural study techniques universit vincent word workshop
http://acl.ldc.upenn.edu/P/P06/P06-1071.pdf	70	A Progressive Feature Selection Algorithm for Ultra Large Feature Spaces	adam adwait alexander algorithm algorithms american analysis andreas annals annotation annual applied approach arbor arpa association attachment bank barbara barcelona berger berkeley best boosting boundaries byrne california canada carnegie channel chapter charmichael charniak cheating chen chunking cmucs coarse comparison computational conditional conference confidencerated conll consortium conversational darby darroch data database della detection different discriminative disfluencies disfluent distribution dustin ears edit edited effective elizabeth empirical entropy estimation eugene even exploring fast feature features fields from fuliang gaussian generalized goodman harper hauke hillard however http human icassp identification identifying ieee improve improved improvement incremental inducing intelligence international isca iterative izhak japan jeff jeffrey jeremy john johnson joshua kahn koeling labeled labeling lafferty language learning lease leslie lide linear linguistic linguistics lisbon machine malouf marcus mari mark mary mathematical matthew maxent maximum meeting mellon metadata methods model modeling models natural nature noisy north oracle ostendorf over parameter parsing pattern peskin phil philadelphia phrase pietra pittsburgh plainsboro portugal predictions preliminaries prepositional prior proc proceedings processing program projects prosodically prosody random randomsplit ratcliff ratnaparkhi recognition references region regions regularization relaxed repairs report reranking research results reynar riezler robert ronald rosenfeld roukos salim sapporo scaling schapire schmidt score selected selection sentence sentences sequential shafran shattuckhufnagel shriberg simple singer smoothing spain specification speech spontaneous springer stanley statistical statistics stefan stefanie stephen stolcke structural supervised table tagbased taibei taiwan technical technologies technology theory thesis this tofine tomalin transactions transcribed understanding university upenn using uweetr vancouver vapnik various vasserman version vincent vladimir volume washington weakly weng william with wong woodland workshop yang yaqian yoram york zhang zhou
http://acl.ldc.upenn.edu/P/P06/P06-2090.pdf	236	Implementing a Characterization of Genre for Automatic Genre Identification of Web Pages	across addressing american analysing applications artificial automatic bathia beghtol biber blood boese bulletin business cambridge characteristics cikm clarifying classification collection communication concept concepts conference creating crowston december decision detection digital dillon document domains duda eacl effects emergent evolution experience facetted first forum framework genre genres germany gushrowski hawaii history home howe inform information intelligence intern issues jasis joho kessler kwasnik language large london longman making multidimensionality navigating norwood numberg overview page personal perspective pocket press proc professional prospector rebecca reboh references registers reitman reproduced sanderson science settings sigir society speech spirit styles system technology text types uniquely university variations weblogs williams worldwide writing
http://acl.ldc.upenn.edu/P/P06/P06-4001.pdf	287	FAST ­ An Automatic Generation System for Grammar Tests	about adapting against aided algorithmically american among amount applications applied arbor article articles assistance assisted automatic awets barrons based blindly book books brookline building calico cambridge canada categories center chall challenging characteristic choice classified clearinghouse cloze composing computer computers concerning conference coniam constructed converted corpus covering dale data deemed designed detection digest direct distribution each edmonton education educational encyclopedia english eric error evaluated evaluation experiment extracted finally first formula freeman frequency from generation grammar grammatical hoshino http huang iccai icce indicate information inquiry instruction international into issue items journal july language languages large larsen lexical linguistics matched method michigan minor mitkov multiple myths naacl nakagawa natural needed nine number online only organized patterns performance preliminary prepare proceedings processing professor program proposed question questions readability realtime references regarded related results retrieved revision revisited rules satisfactory second selected selfcontained sentence sentences series seven sharpe simulated stems students study system tagged tasks teaching tesol test testing tests then toefl total traditional transformed using verb voice vtaide wang washington webbased websites were while wikipedia with word workshop worthy written
http://acl.ldc.upenn.edu/P/P06/P06-2042.pdf	188	Detection of Quotations and Inserted Clauses and its Application to Dependency Structure Analysis in Spontaneous Japanese	analysis annotation annual applying association automatic automatically boundaries boundary building chunking clause clauses coling conclusion context contributed corpus dependency dependencystructure described detecting detection estimated evaluation experiment experiments found furui future hamabe hanae hideki hitoshi identification ieee improvement improving inserted investigate isahara isca iwpt japanese kashioka katsuya kawahara kazuya kikuo kiyotaka koiso kudo kumano kurohashi language lrec machines maekawa makoto maruyama masaki matsumoto meeting method model modification murata naacl nagao natural nineth nlprs pages paper parsed parsing plan posterior problems proceeding proceedings processing quotations recognition references results robustness rules ryoji sadao sadaoki satoshi sekine sentence sentences shitaoka showed solve speech spontaneous structure support system tadashi takanashi takehiko taku tanaka tatsuya that their this uchimoto using vector when while with workshop yuji
http://acl.ldc.upenn.edu/P/P06/P06-1017.pdf	16	Relation Extraction Using Label Propagation Based Semi-supervised Learning	acmdl advances agichtein among annual aone applied april association barcelona based belkin blum brin brown cald charniak chawla cikm classification collections combining computational computer conference corpora culotta data database department dependency development digital disambiguation discovering divergence document emnlp empirical entities entropy exploring extending extract extracting extraction features fields friedman from functions gaussian ghahramani graph gravano grishman guodong harmonic hasegawa ieee infomation information inspired international jian july kambhatla kernel kernels knowledge label labeled lafferty language large learning lexical libraries linguistics machine management manifold maximization maximum measures meeting methods miller mincuts models named natural neural niyogi novel pages parser parsing partially patterns philadelphia plain proceeding proceedings processing propagation ramshaw randomized reddy references relation relations report research retrieval richardella rivaling rwebangira science seattle sekine semantic semi sense sequential sergey shannon sigir slonim snowball soresen spain statistical structure supervised syntactic systems tech technical technology text theory tishby transactions tree university unlabeled unsupervised using various washington weakly webdb weischedel wide with word workshop world xiaojin yarowsky zelenko zhang zhou zoubin
http://acl.ldc.upenn.edu/P/P06/P06-2013.pdf	159	An Empirical Study of Chinese Chunking	abney andrew another antal anthony applying approach approaches armstrong association base based basephrases beijing berwick bosch bracketing brill buchholz byoung cardie carol case changning chen cheng chia china chinese chunk chunking chunks chunyu church cicling city claire coling combination combining compared computation computational computers conclusions conditional conducted conference conll corpora crafted crfs daelemans data david dordrecht driven editors empirical entity entropy eric erik error fang fernando fields fien first four guidelines guodong halteren hammerton hand hans heng hongqiao http huang hung hybrid icml identification ieee ijcnlp improving independent international introduction issue jakub james jersey jian jianfeng jianmin jingbo jmlr john jonathan jorn journal kachites kenneth kluwer kroch kudo labeling lafferty lance language large learner learning lian linguistics lisbin lisbon machine machines mallet marcus matsumoto maximum mccallum memory meulder mexico miles mitch models muyun naacl named natural nature nedellec nianwen nlpke osborne pages paper park parsing part penn pennsylvania pereira performance phrase plus portugal principle probabilistic proceedings processing psycholinguistics qiang qing ramshaw random recognition references report robert robust rocling rules sabine sang second segmenting seong sequence shallow shared shih shizhe sighan sigmoid sloot somerset spec special speech springer statistical statistics steven study sujian support susan svms system tagging taku task technical tenny text theory third this tianshun tiejun tilburg timbl tjong tongguan toolkit transductive transformation treebank tsai type tzong umass university using vapnik vector veenstra verlag very walter webster with wordclass workshop xiaozhong yang yarovsky yongmei york yuji yuqi zavrel zhang zhao zhifeng zhou
http://acl.ldc.upenn.edu/P/P06/P06-1058.pdf	57	An Equivalent Pseudoword Solution to Chinese Word Sense Disambiguation	accent ambiguity annual application approach artificial association automatic based bootstrapping brown category chan chew china chinese companion computational conference corpora david decision della disambiguation discrimination dong empirical evaluations exploiting flairs florida french hearst hinrich hong hwee intelligence international iterative joint jscl label languages large learning lexical linguistics lists lrec marti meeting mercer method methods mihalcea moldovan naacl nakov pages papers parallel peter pietra preslav proc proceedings propagation pseudowords rada references research resolution resources restoration rivaling robert schutze semi seng sense short society spanish stephen study supervised symposium tagged texts track unsupervised using vincent volume wang word xiaojie yarowsky ying zheng
http://acl.ldc.upenn.edu/P/P06/P06-2067.pdf	213	Parsing and Subcategorization Data	according acquisition affected alternation alternations american annotated annual applications applied argument association automatic based behavior bikel brent brew briscoe brown building calibrating cambridge carroll chapter charniak choice class classes classification clustering collin collins colloquium comlex computation computational conference corpora corpus czech development diathesis dictionary disambiguation disfluency distribution driven empirical engel engineering english entropy extraction features frames frequency from germany godefrey grammar grishman head henderson holliman icassp identification informative inspired interdisciplinary international intricacies joanis johnson jurafsky korhonen labeling language lapata large learning lexical lexicon linguistics macleod manning manuscripts marcinkiewicz marcus maximum mcdaniel meeting merlo meryers methods midwest minnen model models morphological motivation natural necessity north pages palmer parser parsing part pearce penn pennsylvania placement predictive priors probabilistic proceedings processing punyakanok ratnaparkhi references representation research roark robust roland role roth saarbracken sarkar schulte semantic semantically speech spoken statistical stevenson structure subcategorization switchboard syntactic syntax tagging telephone thesis treebank university unsupervised using verb verbs walde workshop zeman
http://acl.ldc.upenn.edu/P/P06/P06-1101.pdf	100	Semantic Taxonomy Induction from Heterogenous Evidence	achieving acquisition allowed applications artificial automatic between brown buitelaar caraballo cimiano distance each edges evaluation first found from frontiers hierarchy highest hypernym hyponym intelligence labeled learning magnini maximum methods model noun only ontology references score sense text that thesis university used volume wordnet
http://acl.ldc.upenn.edu/P/P06/P06-2073.pdf	219	Segmented and unsegmented dialogue-act annotation with statistical dialogue models	aaai abella abeth acceptance access acoustics acquisition action acts afternoon agents alcacer alexandersson alicia allen annotation annual answer arrive assessment association aust automated automatic baker bates before bened bernsen bianka biasca birte blat boston buschbeck call canada candace center chapter classification classifiers coccaro coders coding cognitive cohen colorado communication communicative computational computer conf conference conversational core corpus cristian current customer damsl december development devillers dfki dialog dialogs dialogue diane dihana directions discourse discriminative draft dybkjaer edinburgh edition editors edmonton eighth elis environments errors essdykema estimation european evaluating example fall fare fares features fifth figure framework fraser fujinami function germany gmbh godfrey gorin granell greece gruyter harbeck hardy help hepple hilda holliman hour human humans icassp ieee informaticas information institute integrated interaction interactive international interpolation intra isle james jersey jornadas july june jurafsky kamm kipp kirk koch kompe kuppevelt labelling lamel language laurence layer layla level levit linguistics litman lleida lori machines maier malaga management manual march marilyn mark mart martin mate mcdaniel meeting meetings melanie meteer methods michael modal model modelling models mouton multi multilingual multiparty naacl nacional nick niels niemann norbert notes noth november oerder pages paradise parameters patras philadelphia philip philips philosophical phoenix pittsburgh probabilistic proc proceedings processing processings produced programa prosodic question recognition references reithinger report reports research rhodes riccardi ries rosset royal saarbrucken scheme schmitz science scotland second segmentation seguimiento seide series service seven several shallow shriberg siegel signal smith society somerset sophie spain specom speech spoken spontaneous springer steinbiss stephan stolcke strzalkowski swbd switchboard symposium system systems tagging taylor technical technology tecnolog telephone text that thirty thursday time times timetable tomek torres train trains trans tsutomu understanding university ursu using utterance varona verbmobil volume wahlster walker want warnke webb wilks with wolf wolfgang workbench working workshop wright young
http://acl.ldc.upenn.edu/P/P06/P06-2110.pdf	256	Word Vectors and Two Kinds of Similarity	aitchison akio albert amano american analysis associations barabasi basil behavior blackwell blocks building burgess computers conceptual curt deerwester dumais edition editors emer from furnas gence george goitokusei harshman hayashi hiromi ikehara indexing information instruments introduction japanese jean journal kentaro kimihisa kondo landauer language laszlo latent lexical lexicon masahiro meaning memory mental methods mind miyazaki model modeling nakaiwa networks nihongo ogura ometry ooyama oxford peter press properties random rdenfors references reka research richard sanseido satoru satoshi scaling science scott semantic shigeaki shirai simple society spaces susan thomas thought tokyo with words yokoo yoshifumi yoshihiko
http://acl.ldc.upenn.edu/P/P06/P06-2096.pdf	242	Adding Syntax to Dynamic Programming for Aligning Comparable Texts for the Generation of Paraphrases	acids alignment analysis anders approach barzilay biological bootstrapping buchholz cambridge choice chunklink durbin eddy edmonton emnlp graeme html http krogh learning lexical lillian mitchison models multiple naacl nucleic paraphrase perl philadelphia press probabilistic proceedings proteins readme references regina richard sabine script sean sequence university unsupervised using
http://acl.ldc.upenn.edu/P/P06/P06-2016.pdf	162	Techniques to incorporate the benefits of a Hierarchy in a modified hidden Markov model	algorithm automatic based bikel borkar brill case deshmukh driven error into language learning learns machine name natural proceedings processing records references sarawagi schwartz segmentation sigmod structured text that transformation weischedel what
http://acl.ldc.upenn.edu/P/P06/P06-2049.pdf	195	Transformation-based Interpretation of Implicit Parallel Structures: Reconstructing the meaning of vice versa and similar linguistic operators	advanced aied annotated artificial aspects australia benzmuller company conference corpus dialogs dialogues dordrecht draft education europarl evaluation experiment fiedler gabsdil hajicova horacek intelligence international karagjosova koehn korbayova kruijff language lisbon lrec machine mathematical mathematics meaning multilingual netherlands pages panevova pinkal potugal pragmatic proceedings proving publishing references reidel resources semantic sentence sgall siekmann supplementary sydney technologies theorem translation tsovaltzi tutorial unpublished viii wizard wolska workshop
http://acl.ldc.upenn.edu/P/P06/P06-2078.pdf	224	An Automatic Method for Summary Evaluation Using Multiple Evaluation Results by a Manual Method	anlp automatic challenge chinese comparison donaway drummey evaluation fukushima hidetsugu japanese kevin laura manabu mather measures meeting naacl nanba notes ntcir okumura part proceedings produced rankings references research retrieval robert second summarization takahiro text working workshop
http://acl.ldc.upenn.edu/P/P06/P06-1022.pdf	21	Dependency Parsing of Japanese Spoken Monologue Based on Clause Boundaries	agarwal analysis annotation approach based bigram boggess charniak cocosda collins conjunct core delmonte dependencies dependency disruptions emnlp entropy eurospeech framework from fujio huang icslp icslt indentification inspired japanese kashioka lenhart lexical lexicalized mark maruyama matsumoto maximum model monologue naacl other pages parser parsing proc punctuation references repairs schubert segmentation semantic simple speech spontaneous statistical statistics structure syntactic unit useful zweig
http://acl.ldc.upenn.edu/P/P06/P06-2085.pdf	231	Using Machine Learning to Explore Human Multimodal Clarification Strategies	algorithm attributes bart based bayesian class classification classifiers cohen combined communicator conf continuous continuousvalued correlation daelemans data dialogue discrete discretization distributions ecml effective estimating fast fayyad feature fien from george georgila hall henderson hoste hybrid icml ijcai induction interaction irani james john kallirroi kaufmann keki knowledge langley language learning lemon machine mark meulder morgan multiinterval naudts numeric oliver optimization parameter policies practical proc proceedings reasoning references reinforcement rule selection supervised systems usama veronique walter william workshop
http://acl.ldc.upenn.edu/P/P06/P06-3011.pdf	282	On2L - A Framework for Incremental Ontology Learning in Spoken Dialog Systems	acquisition algorithm algorithms andreas annotation argument arndt automatic berenike charles cimiano claire class classification clifford conference context cormen daniel david dijkstra domains driven eberhart edition entity faulhaber faure foundational from genoa gimme gunter hill hitzler introduction italy knowledge ladwig learning leiserson loos lrec machine malaka mcgraw multiple named nedellec oberle ontolex ontology open pages pankow pascal philipp porzel predicate press proceedings project rainer references report rivest robert ronald rudi second section semantic smartweb staab steffen stein structures studer system technical texts thomas towards understanding unknown using wide with workshop world
http://acl.ldc.upenn.edu/P/P06/P06-1027.pdf	26	Semi-Supervised Conditional Random Fields for Improved Sequence Segmentation and Labeling	abney accelerated advances algorithm altun analysis annual artificial association belkin bengio bioinformatics biomedical biotagger blum bousquet boyd calculation cambridge castelli celeux certainty chapelle classification clustering cohen combining computational conditional conference convex corduneanu cover cozman crfs data dependent deriving descent directed disambiguation documents duda editors elements entity entropy everson field fields first form from functions gaussian gene ghahramani global govaert gradient grandvalet graph harmonic hart held http huang identifying ieee information intelligence international jaakkola john labeled labeling lafferty language learning lerman linguistics local machine mallet margin maximum mcallester mccallum mcdonald meeting mentions meta methods minimization mitchell mixing modeling models murphy national navin neural nigam nocedal note numerical obtained optimization parameter partitioning pattern pereira performance pfedc press probabilistic proceedings processing protein random recognition references regularization relative rezek risks rivaling roberts samples scene schmidt scholkopf schraudolph seas segmenting semi semisupervised sense sequence sets show simple sistency software some sons springer sryantm statistics stochastic structured supervised suppl syntactic systems table tagger text that theory therefore this thomas thrun toolkit topic trained training trans twentieth umass understanding university unknown unlabeled unsupervised upenn using value vandenberghe variables vector versions vishwanathan weston wiley wish with word workshop wright written yarowsky yields zhou zien
http://acl.ldc.upenn.edu/P/P06/P06-4008.pdf	294	K-QARD: A Practical Korean Question Answering Framework for Restricted Domain	about adams adaptation addition airs akiba also annotated answer answering application asia automatically baatz based broadcast built chung classifier collection computer concepts consisting contextual crabbe dependency dictionary domain domains driven each entity evaluate examples experiments extracted focus frames framework francisco from fujii graduate half ieee information ishikawa itou kaufmann language languages learned learning lecture lexical local machine martin module morgan note ntcir ontology order pages parser pattern proceedings programs publishers qard question questions quinlan recognition references restricted retrieval rewrote ross rules science semantic several site song speech speechacts spoken symposium system systems target templates third title took towards traffic training unlexicalized using variable weather week which word workshop yankelovich
http://acl.ldc.upenn.edu/P/P06/P06-2074.pdf	220	ARE: Instance Splitting Strategies for Dependency Relation-based Information Extraction	aaai acquisition adaptive algorithm alignment allowed alone also although ambiguity anchor anchors another answering applicable applications applied approach approaches aseltine automatically average based because believe between biocomputing biological bootstrapping cascading case categories category chieu chua chunk ciravegna clue coling comparison conceptual conclusion context cooccurrence crystal cues culotta current data definitional demetriou dempster dependency dictionary differences direction domain dominates each entities entity entropy enzyme existing explore extracted extraction fall first fisher free from future gaizuskas generalization generalized generating generation generic given grishman hard however humphreys ieee ijcai important improvement incomplete inducing induction information instances integrate interactions into investigate journal kaufmann kernels knowledge laird language learning lehnert less likelihood limitations linguistic local machine main management maslennikov matching maximum meaningful methods missing mistakes models moldovan more named names observe occurrence ontology outperform over overall pacific paraphrasing parsing pattern patterns performance plan primary probabilistic problems proc process promotion proposed protein provide qualifier question reason reasoning recognition references relation relations require resolve results riloff role roth royal rubin rule rules science second semantic semi sense serious sigir similar simple society soderland soft sorensen specific split splitting stable state statistical strategy structured structures succession supervised symposium syntactic system table tackle tagger task tend terminologies terrorism text that them there therefore these third this training transactions transduction trec tree trends types unsupervised untagged used using utilize very weakly with word work xiao yang yangarber zhou
http://acl.ldc.upenn.edu/P/P06/P06-1034.pdf	33	Learning to Generate Naturalistic Utterances Using Reviews in Spoken Dialogue Systems	about adjectival alignment alistair amazing answering appendix applications applied approach atmosphere awesome barzilay based beautiful benoit best bing bontcheva bootstrapping busy capacious casual categorization choice christiane class coherence comfortable communication comparative contemporary cool corpus cozy cunningham customer data database decent dekang delightful dependency descriptions development dialogue diana diane discovery driven eclectic edinburgh electronic ellen emnlp engineering enjoyable environment evaluation excellent exploiting extracting extremely fancy fantastic fast fellbaum flair foster framework friendly from gate generating generation good graphical great hamish hirschberg humiliating identifying idyllic inference interesting interior intimate intonationally johanna julia just kalina kathleen knott language lavoie learning lemon lexical lillian litman local lovely mappings marvelous mary maynard mckeown mellow methodology michael mining minipar minqing moore motivating multiple multiplesequence naacl natural neat never nice oliver outstanding overall overwhelmingly owen pages pampering pang pantel parallel paraphrase paraphrases parsing patrick peaceful phenomenal phrases pleasant portable press pretty proc question quite rambow rating realizer really reasonable recommended references regina relation relations relationships relaxing respect reviews robust rules scalar scales seeing sentiment sequence simple single special speech spoken stars summarizing systems tablan tailored talk tasty text thesis thoroughly tools totally trendy truly typical ultimate unbelievably unique university unsupervised using valentin value valued very warm white with wonderful wordnet workshop
http://acl.ldc.upenn.edu/P/P06/P06-2057.pdf	203	A FrameNet-based Semantic Role Labeler for Swedish	baker berkeley canada charles coling collin fillmore framenet john lowe montr pages proceedings project references
http://acl.ldc.upenn.edu/P/P06/P06-1002.pdf	1	Going Beyond AER: An Extensive Analysis of Word Alignments and Their Impact on MT	aachen abraham aligned aligner aligning alignment alignments alon among amta approach arabic automatic ayan banerjee based beam bilingual bleu bonnie brown burch callison chiang chris christof christoph coling combining comparison computational corpora correlation cyril daniel david decoder della discriminative dorr emnlp english entropy equivalence eric error estimation evaluation extrinsic factorisation framework franz gaussier giza goutte hermann hierarchical hltnaacl human improved intrinsic ittycheriah jing joint judgments julien kenji kishore klein koehn lacoste lavie linguistics machine march marcu matching mathematics matrix maximum measures melamed mercer meteor method metric miles minimum model models monz moore necip networks neural neuralign osborne pages papineni parallel parameter peter pharaoh philipp phrase pietra probability proceedings rate references report robert roukos rwth salim search sentence simon stanjeev statistical stefan stephan summarization systematic talbot taskar technical technology tillmann todd training translation translational university using various vogel ward william with wong word words workshop yamada
http://acl.ldc.upenn.edu/P/P06/P06-1041.pdf	40	Hybrid Parsing: Using Probabilistic Models as Predictors for a Symbolic Parser	almost amit analysis anlp annual anytime applied approach aravind arbor artificial attachment automatic bangalore based benefit berlin brants brigitte charniak christiansen clark coling collins computational computerlinguistik conf conference constraint constraints cordula coverage curran daum defeasible dependency dipper driven dubey editors entropyinspired eugene evaluation fails foth george german germany hans hansen head hendriks importance ingo intelligence iwpt james joshi kilian kramp krenn language lecture lexicalization lezius linguistic linguistics lisbon lrec maximum meeting menzel michael models naacl natural negra negraannotationsschema notes ofspeech pages parser parsing part pennsylvania philadephia phrase portugal preis proc proceedings processing project properties references report resources roland rule saarbrucken saarlandes sabine schroder seattle silvia sixth skadhauge skut smith smoothing solving sozopol springer srinivas statistical stefanie stephen stochastic suffix supertagging sydney tagger technique technologies text theories thesis thorsten tiger transformation treebank treebanks trees universitat university unrestricted uszkoreit verlag villadsen volume what when wide with wojciech wolfgang workshop
http://acl.ldc.upenn.edu/P/P06/P06-1129.pdf	128	Exploring Distributional Similarity Based Models for Query Spelling Correction	accuracy acquisition adam after ahmad andrew annals annual applying approach automatic automatically based basically beger better binary brill capable changes channel church clustering codes coling collective communication computation computer computing concluded conclusions context correcting correction cucerzan current curve dagan damerau darroch data dekang deletions della detection different disambiguation discriminative distance distilled distributional doklady edit emnlp enough ensure entropy eric error errors essen evaluate evidence explained exploiting exploits farooq feature fernando figure franz fred from functions future gale generalized golding grzegorz hanging have hermann icassp icml impact improved information insertions iterative karen kenneth kernighan knowledge kondrak kristina kukich language lawrence learn learning levenshtein lidia lillian linear linguistics logs long machine magazine maimum management mangu mark mathematical maximum mayes measures meeting mercer metaphone methods model modeling models moore motivation natural noisy novel number occurrence only pages pereira performed peter philips physice pietra presented probabilistic probability proceedings process processing program pronunciation proposed query ratcliff reaches recall references results retrieval reversals ristad robert roth rule samples scaling scoring search sense sensitive sets shows silviu similar similarity size smoothing soviet spelling statistical statistics steinbiss stephen stochastic string study subtle sufficient surveys take task technique techniques text that there therefore toutanova train trained training translation users vincent volker volume william winnow with word words work yianilos
http://acl.ldc.upenn.edu/P/P06/P06-1104.pdf	103	A Composite Kernel to Extract Relations between Entities with both Flat and Structured Features	accuracy accurate achieves acknowledgements acyclic addition alessandro algorithm algorithms also anonymous aone automatic based basili benefit benefiting bestreported beyond bunescu calculations california cambridge cammisa captured capturing categorization charniak classification classify collins combine combining comparable composite conclusion content convolution corpus could cristianini cruz culotta data demonstrate dependency design designed detection directed discrete done duffy ecml effective embedded emnlp engineering entity entropy examples explore exploring extension extensive extract extracting extraction feature features find fine first flat flexible from functions future graph great grishman haussler have head help hierarchical hirao http iaui icml imbalance immediate important improve including individual information instance integrated invaluable issues joachims journal kambhatla kernel kernels knowledge language learning lexical like lkopf lodhi machine maeda many maximum measure methods miller models mooney more moschitti most naacl natural need nice nips nist novel optimization other outperforms paper parse parsing particularly path performance portions poster press previous projects properties ramshaw references regularization related relation relations relevant report representation research reviewers richardella santa sasaki saunders selecting semantic shallow shawe shortest shows similarity smola solve sorensen sparseness statistical string structured structures study such suggestions support suzuki syntactic system taylor technical text thank that their therefore this three toolkits training tree trees tuning university upenn useful using various vecor very viewpoint watkins weischedel well which will with without wordnet work would zelenko zhang zhao zhou
http://acl.ldc.upenn.edu/P/P06/P06-1030.pdf	29	Automated Japanese Essay Scoring System based on Articles Written by Experts	ability achievement aera algorithm american analysis annual applications applied assessment associates association automated available baayen based bejar bennet bereiter berry betsy board braden bulletin burstein cambridge chase chodorow choice communication comparing composition comprehensive computational computations computer computers conference constant context cooccurrences cooper council cross crossdisciplinary data deerwester department differences disciplinary document duff dumais edmedia education educational effects elliot english erater ericae erlbaum errors essay essays european eurospeech examination examinations expectations feature finding foltz foreword fowles from fuculty furnas grades grading grammar greb greenbaum grimes handwriting harder harris harshman here hillsdale html http hughes human humanities hybrid identification impact improved indexing information inoue instructions intellimetric intended interaction international isbn ishioka issues iwanami japanese jelinek journal kameda keeling keith knuth kukich laham landauer language large larrabee latent lawrence leech lewis lexical liang linguistics literary logical longman maekawa marshall math mathematical matrix measurement measures meeting meyer models nagao national natural ncme neatness noya number online only orleans page papers perspective poggio powers practice press princeton problem processing profile psychological psychology quality questions quirk reduce references relevant report research reswrit retrieval review richness roberts rudner sangyo scale science scientific scorers scoring semantic series service shermis shotton singular society software sparse speech stancs stanford statistical statistics struggle student study supercomputer svartvik symposium taira technique technology test testing tests theassessmentofwriting tokyo tosho training trait trans trigrams tuck tweedie university using validity value variable variables vocabulary watanabe wolff words writing yule
http://acl.ldc.upenn.edu/P/P06/P06-2017.pdf	163	Analysis and Synthesis of the Distribution of Consonants over Languages: A Complex Network Approach	abry academic acoustically adamic albert allgemeine amsterdam aphasie architecture auditory barabasi bart boer boersma bulmer cancho chomsky clements collaboration commerce complex congr congressus consonant crosslinguistic cuny diversity doctoral electronic elsevier emergence english feature features ferrer flemming functional gomes graph graphics greenberg grossman hague halle harper hierarchies hinskens holland huberman icphs inventories jakobson journal kindersprache known language languages lautgesetze linguistics london markets mathematics mechanics modern modification mouton nature networks numerantium optimal organisation pages paper pattern patterns phonetics phonological phonology physica physics portion principles quarterly random reference references relations representations reprinted reviews routledge santa scaling science sciences segmental selected self sole sound special statistical statistics study symposium systems theory thesis triangle tsang universal universals university uppsala vasconcelos vowel weijer well wide with working world writings york
http://acl.ldc.upenn.edu/P/P06/P06-2120.pdf	266	Stochastic Discourse Modeling in Spoken Dialogue Systems Using Semantic Dependency Graphs	abeille allen anne annotated automatic baker berkeley building byron chang chen coling computation computational computer conversational corpora criteria dependency design dong editor ferguson fillmore framenet galescu gildea hacioglu hownet huang human immplementation interaction issues jurafsky kluwer labeling language learning linguistics lowe machines magazine martin meaning modeling parsing pradhan proceedings project publishing references report representational roles scientific semantic shallow sinica stent structure support suzuki syntactically technical towards treebank unsupervised using vector ward world
http://acl.ldc.upenn.edu/P/P06/P06-2083.pdf	229	A Term Recognition Approach to Acronym Recognition	abbreviation abbreviations acids adar algorithm alice american amia ananiadou artech association automatic bioinformatics biology biomedical biomedicine bodenreider carol chang dictionary domain editors evaluating expression extract extracting extraction eytan frantzi friedman from george goran hinrich hiroko hisamitsu hongfang house hripcsak independent informatics integrating jeffrey john johnson journal katerina knowledge language lexical lyuda management mcnaught medical medline method mining multi natural nenadic niwa nucleic olivier pages parenthetical processing references research robust sarad schutze shagina simple sophia source stephen symposium system takagi term terminology terms text toru toshihisa umls unified useful value word yoshiki
http://acl.ldc.upenn.edu/P/P06/P06-2071.pdf	217	Discriminating image senses by clustering with multimodal features	annotated annotating barnard challenges corpora disambiguation forsyth frontiers iccv images learning linguistically loeff pictures references semantics sense words workshop
http://acl.ldc.upenn.edu/P/P06/P06-1072.pdf	71	Annealing Structural Bias in Multilingual Weighted Grammar Induction	abeille acoustical acquiring across afonso algorithm alshawi america analysis annealing annotated annotation applications apply approach appropriate around assumes atalay automata automatic automatically automaton available baker based bears best bias bick bilexical bilingual boot bootstrapping bosque bottleneck brants brill brown buchholz building bulgarian bultreebank charniak chinese chiou choice chooses chosen clark classification clustering collins come comparing compression computation computational compute computed computes conll connected consist consists constant constituency constituentcontext constraints context continuation contrastive conventions converges corpora corpus correspondences counts crammer crim criteria cubic darpa data datasets della dempster dependency described design deterministic development diagonal different difficult dipper directly disambiguation distribution distributional done drawn dynamic each eacl editor editors efficient eisner either elidan emnlp engineering english envision estimating estimation europa event events every exact example expectations expected experimental experiments exploiting extrinsic factored fast figures first five floresta following follows free friedman from future gaussian general generative german gold grammar grammars grammatical guiding haber hakkani hansen hard have head hence here hidden highest hindle hockenmaier hpsg hpsgbased hyperparameter hyperparameters hypothesis hypothesized ideally ieee ijcai implementation improved including incomplete induction inference information initialization initializer initializers inside inspiration iterations iterative iwpt journal karakos kinds klein kluwer kouylekov laird language languages large learning length lexicalised lezius like likelihood linc lincom linear linguistic linguistics lrec machine mandarin manning many marcinkiewicz marcus margin marsi mathematics maximum mcdonald mcenery mean means measure measures mention mercer merit methods metusabanci minimal model models multilingual nakano natural need networks neural next nips noun objective observable obstacle occurs oflazer online only optimization oracle organizers osborne osenova other outside over pages palmer paper parameter parser parsers parsing passed penn pereira performance performs perhaps phrasal phrase pietra plausibility popova portion portuguese practical precision predicateargument present presented press prior probabilistic probabilities problems proc proceeds process programming proper punctuation quite rainbow rayson recall recent recognition recursive references regression related relative renormalizes report representations required research results rivaling rose royal rubin ruhlen santorini santos sarkar satta scaled schedules scheme scored scores seed select selected selection sense sensitivity sentence sentences sequences serves setting settings setup shared similarity simov simple simultaneously since single sinta small smith smoothing society soft some special speech standard started statistical steedman step stripped structural structure structures successful such summed supervised syntactic syntactically syntax task technique techniques test thank that then theories these this three tica tidy tiger tiling time tolerance trainable trained training trajectories translation tree treebank treebanks trees trials turkish ueda under uniform unlabeled unsupervised until useful using usual value values variable weakness weighted when where which whichever will wilson with within without word words work workshop world worst would yarowsky yleft yright zero
http://acl.ldc.upenn.edu/P/P06/P06-1024.pdf	23	Learning More Effective Dialogue Strategies Using Limited Dialogue Move Features	architecture austin automatic cambridge clark description dialogue dipper discourse eckert esther evaluation ewan formalisation frampton herbert ieee ijcai information johan john klein knowledge language last learning lemon levin matthew modeling oliver oxford pieraccini practical press reasoning recognition references reinforcement roberto sapporo sigdial speech spoken state strategies system systems tetsushi things understanding university update user using weiland with words workshop
http://acl.ldc.upenn.edu/P/P06/P06-1078.pdf	77	Incorporating speech recognition confidence into discriminative named entity recognition of speech data	allen andrew approach archives bechet beno borthwick chieu chiori communication composition conll context continuous data detecting dialogue dilek discriminative emnlp entities entity entropy eurospeech extracting extraction fast favre finite frederic from gorin hakkani help hori horlock hwee icslp improving initiative james jeremy king large lattices leong maximum methods million minami mixed named nocera pages pascal proc recognition references robust simon speech spoken spontaneous state takaaki thesis transducers university vocabulary volume weighted with word wright yasuhiro york
http://acl.ldc.upenn.edu/P/P06/P06-1066.pdf	65	Maximum Entropy Based Phrase Reordering Model for Statistical Machine Translation	algorithm ashish based bilingual block boston budapest chiang christoph classification computational considerations corpora david dekai eamt error grammars hierarchical hungary information inversion linguistics localized machine maximum minimum model mutual naacl orientation pages parallel parsing phrase polynomial prediction proceedings references statistical stephan stochastic tillmann time tong training transduction translation venugopal vogel zhang
http://acl.ldc.upenn.edu/P/P06/P06-2113.pdf	259	Combining Statistical and Knowledge-based Spoken Language Understanding in Conditional Models	accelerated accuracy acoustics advantage algorithm algorithms also applications approach approximation atis audio backward bahl based beneficial benefit biggest brown chou classification collect collins composite conclusions conditional conference counts criterion crossvalidation darpa darrell data demonstrated deng descent directly discriminative discussed domain each ease emnlp error estimation eurospeech evaluation expected experience experiments faster feature features fewer fields forward framework generative generatively given gunawardana have help hidden icml ieee improved improvement information integrates international into introduced introduction iteration iterations juang knowledge kushner label labeling lafferty language learning like lisbon magazine mahajan markov maximum mccallum meta methods minimum model modeling models more mostly much mutual natural need nips nocedal numerical object observation occurrence only optimization over overall overlapping paper parameters perceptron philadelphia phone porting portugal povey price probabilistic processing quattoni random rate recognition reduces reference references replicating required schraudolph segmenting sequence shown signal slot smoothing snowbird specifically speech speed spoken springer statistical still stochastic stopping system than that theory think this times training transactions understanding using utah valley verlag vishwanathan viterbi wang when with woodland workshop wright
http://acl.ldc.upenn.edu/P/P06/P06-3014.pdf	285	Parsing and Subcategorization Data	accurate american association bikel brent brew chapter charniak class coincidence collin collins computation computational conference corpus development disambiguation disfluency driven dunning empirical engel entropy from godefrey grammar head holliman icassp informative inspired intricacies johnson language lapata learning lexical lexicon linguistics maximum mcdaniel methods models natural north pages parser parsing pennsylvania placement priors proceedings processing references research speech statistical statistics surprise switchboard syntax telephone thesis university unsupervised using verb
http://acl.ldc.upenn.edu/P/P06/P06-1074.pdf	73	An Iterative Implicit Feedback Approach to Personalized Search	accurately american categories challenges chengxiang cikm clickthrough communities compass computer craswell data engine evaluation feedback filter granka harman hawking hembrooke implicit information interpreting itwp joachims kritikopoulos mapping meng modeling networks pages pang personalization personalized proceedings queries references result results science search shen sideri sigir society thistlewaite user using xuehua zhai
http://acl.ldc.upenn.edu/P/P06/P06-1010.pdf	9	Named Entity Transliteration with Comparable Corpora	acquisition algorithm aligned alignment alpha ambiguity ambiguous anatomy andrew architecture australia automated automatic ballesteros based bilingual black bootstrapping brin bruce caley carlson caves chang chapter chen chengxiang chinese cognates coling comparable comparing comparison computational computer conference confusion corpora correlation croft cross crosslanguage csli cumby dept development divergence document edition editors edits emnlp engine english entities entity entropy esca extracting extraction fatiha festival figure finding finite fister flournoy foreign franz from fung gale generating graehl hainan handle hill hypertextual identification identifying ieee ijcnlp information integration introduction isdn items iterations iwasaki jenolan july kantor kaufmann knight kruskal language languge large lawrence learning lexical linguistics lisa machine macromolecules martin masatoshi masuichi matching mccarley mcgill mcgraw measures meng method methods mining modern morie multilingual naacl named names networks noisy nonparallel noun number overview page pages pairs parallel pascale pattern peters phil phonemebased phonetic phrasal problem proceedings propagation proper rapp reading recognition references report research resolving retrieval richard robust roscheisen rosen roth roukos sadat salim salton sankoff sanya scale scanned scott search segmentation sequence sergey shannon shih shunsuke sigir snow speech spoken sproat state stochastic string sydney synthesis system systems tanaka tang taylor technical temporal terminology text texts theory third time tracing track transactions translation translations transliteration trec uemura uiuc uiucdcs understanding unsupervised using values vines voorhees warps wong word workshop ying yoon yoshikawa youn zhai zhang
http://acl.ldc.upenn.edu/P/P06/P06-2018.pdf	164	Using Machine-Learning to Assign Function Labels to Parser Output for Spanish	acquisition american approach asia assigning austria available based berger bikel blackwell blaheta bodomo bresnan burke cahill chan chapter charniak chinese computation computational conference dbikel della design development diego donovan engine entropy esslli forst function functional genabith grammar html http human ideas information language lexical lexicalfunctional lingual linguistics march maximum mccarthy multi multilingual natural north oxford pacific paclic pages parallel parsed parser parsing pietra proceedings processing publishers references rochester roher software stat statistical strategies syntax tags technology text treebank unification upenn vienna workshop