http://acl.ldc.upenn.edu/P/P05/
ACL 2005
http://acl.ldc.upenn.edu/P/P05/P05-3024.pdf	132	The Wild Thing!	alphabetical associative bell bentley binary bits cancel case ceil choose code commun communications compressing computer conclude constant constants costs cutrell dependencies development documents dumais finite floor followed francisco frequency gigabytes golomb guess huffman ieee images indexing information into isbn items kaufmann language list managing mcilroy moffat mohri morgan multidimensional near optimal other pereira personal power process publishing quotients recognition recurrence references remainders retrieval riley search searching seen sequence sigir speech spelling splits sqrt state storage store stuff substituting system that this thus trans transducers trees unary used weighted when where witten words work worst zeros
http://acl.ldc.upenn.edu/P/P05/P05-1015.pdf	16	Seeing stars: Exploiting class relationships for sentiment categorization with respect to rating scales	aaai affect aldebaro alex algorithm alison also altering alternatively always analysis andrew answering appendix applications applied approximate artificial assumption atkeson attempted attitude based baseline bernhard best better bigger binary boostexter boosting boykov carnegie case cases categorization category central christopher class classification classifier clauses clearly college computer concern conference confidence conll consider convert correspond cuts datasets decision defense degree desired detecting difference directly discretized discretizing discussion distinction does down editors education emnlp energy error examples experiments exploring facts fast finding first five fixed focus follows found four from function fuzzy goal graph graphs greater hand hatzivassiloglou have higher hiroya hiroyasu holloway hong huettner identifying ieee improvements incorporate increase independently instead intelligence international into issues james janyce jerry journal just kernel klautau koppel label labels large learn learned learning lillian line locally london machine maintain major majority many matsumoto mean mellon minimization minimum modeling moore more nature necessarily needed negative negligible neither neurocolt noticeably olga only operational opinion opinions optimize option order orientation other others outperforms output pages pang path performance pero peter polarity positive positivity preference press proceedings provides questions ramin ratings rebecca references regression relations report research resulting results review reviews rial richard rifkin robert royal ryan schaal schapire schler scholkopf semantic semi sentences sentiment sentimental separating series setting shanahan shivakumar sigir similar simpler singer smola spring springer standard stars started statistical stefan still strong structures stuck study subasic subjectivity summarization supervised support symposium system systems takamura taken technical techniques text than that then theories theory therefore theresa these thesis this three thresholding thresholds thumbs together tong towards tracking train trained transactions treating true turney tuto tweaks typing under university unsupervised using vaithyanathan value vapnik variations vasileios vector veksler version vision vladimir weak weighted well which wiebe wilson with workshop would xiaojin yamada yield yoram yuji yuri zabih
http://acl.ldc.upenn.edu/P/P05/P05-3015.pdf	123	Syntax-based Semi-Supervised Named Entity Tagging	about absolute accuracy adam against algorithm alone andrew annual applying association baluja based best bikel cases classification classifying collins competitive complement computational computer conclusion conll constituency constituent constraints conversion daniel data degradation dempster department dependency difference different documents either entities entity erik examples expected experimented extracted extraction features fien find formed from future general generative given good high higher however incomplete independent introduction journal kamal labeled laird language learning learns lexicalized likelihood linguistics lopez lower machine make manner maximum mccallum meeting methods meulder michael mitchell mittal model models moreover name named nigam nouns pacific paper parsers parsing patterns perform performance pittsburgh proceedings proper proposed rahul ralph rates reasonable reasonably rebecca recognition references report results robert royal rubin rules same sang schwartz science sebastian semi series shared shumeet sigdat since singer size small society solely somewhat statistical suggest sukthankar supervised syntactic systems table task technical test testing text than that their they this three thrun tjong training union university unlabeled unsupervised using very vibhu weischedel well were what when with work yoram
http://acl.ldc.upenn.edu/P/P05/P05-3008.pdf	116	A Voice Enabled Procedure Browser for the International Space Station	aalborg able actions also application approach approaches april architecture backpointer based best browser budapest case categorization chemnitz clarissa clean combination comparing compiling complete conference confirmations contains contrast correct corrections data declarative demo denmark detail dialogue discourse dowding driven each eacl earlier effect enabled engine engineering environment european eurospeech explained features free function functionality germany gorrell grammar grammars have hockey however http hungary idea important include information input interspeech into island issue jeju joachims knight koeling korea language larsson learning lewin like machine machines makes management many microphone milward move much nasa natural object objects open output pages particularly permits practice previous procedure proceedings projects rayner realised recognisers references regulus relating relevant represented representing response restoring robust rule semantics side source sourceforge special speech spoken standard state states straightforwardly study suggests support system systems task text this toolkit track transparent traum treatment trindi typed understanding undo undone undos unification update vector version very voice well whole with within work
http://acl.ldc.upenn.edu/P/P05/P05-1077.pdf	78	Randomized Algorithms and NLP: Using Locality Sensitive Hash Function for High Speed Noun Clustering	accuracy acupressure again algorithms alspector amounts analysis andrei annual another appendix approximate approximation argument association automatic ayurveda balanced banko based biology blood body brill broad broder burberry calisthenics canada cardio categorization cavnar center chanel charikar chemical choosing chosen chowdhury christ church civil classification clustering coling complexity compression computer computing consider considered containment context coverage curran curse dataproblem define dekang detection diego dimensionality dior disaster discovering document documents earthquake economics edmonton efficient electrical engineering environmental eruption estimation fair feldenkrais fendi ferragamo final fingerprinting from function functions general goemans gram gucci hailstorm hanks harvard have heaven here hill hindle hurricane improved indyk information integers introduction jacm japan just kate kingdom kolcz kyoto landslide lauren lexicography lexicon lists louis mapping math maximum mcgill mcgraw means mechanical mitigating modern moens montreal moses motwani mouse mudslide muller mutual near nearest neighbors never nice norms noun pantel parser patrick paucity people permutation philadelphia physics pilates pittsburgh polynomials possibilities possible power prada predicate principar principle problems proceedings programming publications qigong quality rabin ralph random randomization references reflexology removing replica report reprographics research resemblance retrieval robustness rounding salton sample satisfiability scaling science semidefinite senses sequences shiatsu sigkdd signature similar similarity space spade stat stoc structures such symposium table techniques technology texas text theory therapeutic there third thus tidal touch towards trenkle tsunami typhoon unique university unlv using vancouver vegas volcanic vuitton wave weld williamson windstorm with word words would your
http://acl.ldc.upenn.edu/P/P05/P05-1042.pdf	43	A Dynamic Bayesian Framework to Model Context and Memory in Edit Distance Learning: An Application to Pronunciation Classification	acoustics algebraic applications arnborg artificial bahl bartels bilmes channels complexity conf conference corneil corpus decoding deletions development discrete dynamic embeddings finding godfrey graphical holliman icassp ieee information insertions intelligence intl jelinek kaufmann mcdaniel methods models morgan open pages proc proceedings processing proskurowski recognition references research series siam signal software source speech substitutions switchboard system telephone theory time toolkit trans tree triangulating uncertainty volume with zweig
http://acl.ldc.upenn.edu/P/P05/P05-1017.pdf	18	Extracting Semantic Orientations of Words using Spin Model	adam adjectives analysis annealing annual approach association bayesian berger bing carlucci chandler chapter christiane communication computational conference cowie customer data database david della disambiguation discovery distributions domenico donald eighth electronic entropy european fellbaum fifth geman general gibbs glass guthrie hatzivassiloglou ieee image images inoue intelligence international introduction ising journal junichi kathleen knowledge language lexical line linguistics louise machine mathematical maximum mckeown mechanics meeting mining minqing modern natural nishimori orientation oxford pages pattern physical physics pietra predicting press proceedings processing references relaxation restoration review reviews semantic series sigkdd simulated speech spin statistical statistics stephen stochastic stuart summarizing thirty transactions university using vasileios vincent volume wordnet yukito
http://acl.ldc.upenn.edu/P/P05/P05-3002.pdf	110	Accessing GermaNet Data and Computing Semantic Relatedness	academic adapting adjektive alexander american analyse annual application applications applied artificial association august automatic banerjee barcelona based bernhard boston budanitsky cambridge canada canary carroll central chapter christian christiane city claudia cohesion computational computing concepts conceptual cone conference conrath content corpus correcting cream daniel database david decision definition dekang demonstrations deutschen dialogue diana dictionaries disambiguation documentation dordrecht electronic engineering errors eurowordnet evaluate evaluation fellbaum finding first fisseni fourth francisco from geneva germanet gldv global graeme gurevych hans harold helmut hendrik hirst human hundsnurscher india indian information institute intel intelligent international iryna islands jason jiang john joint jones julie july june kluwer koeling kommunikation kunze lang language languages learning lemnitzer lesk lexical ligence linguistics linguistische london lothar lrec machine mass mccarthy measures measuring meeting methods mexico michael michelizzi mobile montreal mutlilingual mysore natural network networks niederlich north ontario palmas part patwardhan pedersen peter phil piek pine predominant press probabilistic proceedings processing publishers readable realword references relatedness relationen representation research resnik resources ressourcen restoring rocling satanjeev schmid schmitz schroder semantic semantik semantischen sense senses siddharth similarity somers spain speech spelling splett spoken sprachtechnologie statistics strube structure studies submitted summarization switzerland systems tagging taiwan tapei taxonomy technology tell text theoretic toronto trees untagged using verlag visualization vossen wagner weeds westdeutscher with word wordnet workshop
http://acl.ldc.upenn.edu/P/P05/P05-3016.pdf	124	Portable Translator Capable of Recognizing Characters on Signboard and Menu Captured by Built-in Camera	achieved against akio akira alex annual application arakawa assistant association automatic background camera character characters chen combining computational computing conclusion conference contrast correction detection digital driving effects error fields from future haritaoglu high hiromi icassp ichi ikehara images incremental information infoscope international into ismail japanese jing kanji kusachi language languages linguistics link masaaki meeting methods model multimedia nagata nakaiwa naoki natural okada other pages pattern preediting proceedings real recall recognition recognizes references road robust satoru satoshi scene scenes shape shirai signboards signs similarity space springer statistical summit suzuki system takeda tetsuya text texture them toward translates translation ubiquitous using variation verlag video viewpoint waibel watanabe with without work world xilin yang yasuhiko yeun ying yokoo yoshihiro yoshinori zhang
http://acl.ldc.upenn.edu/P/P05/P05-1003.pdf	4	Logarithmic Opinion Pools for Conditional Random Fields	accurate acknowledgements acoustics active address advances advantages aggregating aistats algorithms alternative anonymous assessments avoiding baldridge based bayesian between bordley buchholz chunking clark codes cohn colleagues combination comments comparison competitive conclusion conditional conference conll considered constructing conventional cooperative correcting crfs curran data decomposition della designing distribution divergence diverse dynamic each early edinburgh efficiently enhanced ensemble entity entropy error estimation expert experts extraction factors feature features fields firm formula foundation framework from future gillick have heskes hinton hyperparameter icann icml ieee independent inducing induction information intend international introduced introduction investigate issues jointly labeling lafferty language learning lexicons logarithmic lopcrf malouf management many maximum mccallum meulder minka models multiple multiplicative naacl named neural nips opinion optimised osborne outperform overfitting pages pami paper papers parameter parameters parse parsing peng pereira performance pietra pool pools prior probabilistic probability proc processing product provide provides random recognition reduction references regularisation regularised requirement research results reviewers rival rohanimanesh sang scaling schemes science search seen segmenting selecting selection semantics sequence sequences shallow shared show shown signal smith some space speech standard static statistical statistics stephen sutton syntax systems szummer tagger target task terms thank that theoretical these this tjong trained training types under unregularised useful using variety volume weighting weights while wish with work workshop
http://acl.ldc.upenn.edu/P/P05/P05-2022.pdf	102	Using bilingual dependencies to align words in Enlish/French parallel corpora	academic acquisition actes ahrenberg algorithm alignment andersson appariement approach association barbu based bourigault brown clues computational conf conference congr corpora corpus debili della dependencies dependency description ding divergences dordrecht dorr estimation fabre formal gildea improving international jadt kluwer knowledge level lexical linguistic linguistics lite machine mathematics mercer merkel methods mots palmer parallel parameter pendances pietra proceedings processing proposed publishers references rence rfia ronis service simple solution statistical summit syntaxiques text translation trees word zribi
http://acl.ldc.upenn.edu/P/P05/P05-1057.pdf	58	Log-linear Models for Word Alignment	adam alignment annals annual approach association based berger brill brown case chang cherry class colin computational daniel darroch december dekang della dellapietra driven entropy eric error estimation generalized gildea improve iterative japan jason june language learning linear linguistics loosely machine march mathematical mathematics maximum meeting mercer model models natural parameter part peter pietra probability proceedings processing ratcliff references robert sapporo scaling speech statistical statistics stephen study tagging transformation translation tree vincent word
http://acl.ldc.upenn.edu/P/P05/P05-2008.pdf	88	Using Emoticons to reduce Dependency in Machine Learning Techniques for Sentiment Classification	advances analysis annual applications applied artificial association automatic barcelona based budapest burges cambridge categorization cavnar charlotta classification computational conference cuts dave david dependence document down editors education ellen empirical engstrom extraction gallery gram hostile hungary information innovative intelligence international joachims july kernel kushal language large lawrence learning lillian linguistics machine making master meeting messages methods minimum mining natural nevada opinion orientation pages pang peanut pennock pennsylvania peter philadelphia practical press proceedings processing product recognition references retrieval reviews scale scholkopf semantic sentiment sentimental shivakumar smokey smola spain spertus steve subjectivity summarization support symposium techniques text thesis third thumbs topic trenkle turney university unsupervised using vaithyanathan vector vegas wide world
http://acl.ldc.upenn.edu/P/P05/P05-3023.pdf	131	Transonics: A Practical Speech-to-Speech Translator for English-Farsi Medical Dialogues	aaai adaptive although ananthakrishnan appropriate arabic ascii asru based before being belvin bleu calculated called case communication component conclusion conference cooperative corpus creation degree design dialogue different doctor domain enabled encouraging english ettelaie eurospeech evaluation fall farsi from gandhe ganjavi genuine georgiou given have health hein human hurdles ieee implementation indeed interactions interlocutors interviews islands kadambe kind knight language languages lisbon lrec many marcu narayanan narrow neely notes other output overcome overview paper patient patients persian portugal proc proceedings reason recognition references resources robust schemes scores script significant some speech spoken srinivasamurthy standardized still symposium system systems table technology text that there this thomas training transcribed transcription translated translation translator transonics traum truly using very virgin wang while with working
http://acl.ldc.upenn.edu/P/P05/P05-1010.pdf	11	Probabilistic CFG with latent annotations	accurate algorithms based bikel bracketed brendan broad charniak chiang christopher clark coling collins complexity computational context corpora coverage curran daniel david derivation disambiguation driven efficient emnlp entropy eugene extraction feature fernando fitting free frey from generalization goodman grammars head henderson history inclusive inducing inference information insideoutside inspired iwpt jaakkola james japanese jodi johnson joshua jsai khalil klein kodama language latent libin linear linguistic linguistics ltag manning mark matsumoto maximizing maximum metric metrics michael model models moran naacl natural networks nips noisy nondeterministic pages parser parsing partially patrascu pcfg pennsylvania pereira probabilistic proc recovering reestimation references relu report representations schabes sequentially shen simaan specialization statistical stephen syuuji tactic takehito technical terminals thesis tommi tree treebanks trees university unlexicalized using utsuro yuji yves
http://acl.ldc.upenn.edu/P/P05/P05-1048.pdf	49	Word Sense Disambiguation vs. Statistical Machine Translation	acquisition adaboost annual antal application association augmenting barcelona berkeley boosting bosch bottleneck brown california cambridge carpuat carreras clarkson classification computational computer conll data decision decoding dekai della diab disambiguation editors ensemble eurospeech evaluating extraction freund generalization germann greece greeedy international journal july kernel language learning line linguistics machine marine marques meeting mercer methods model modeling mona named padro pages peter philip pietra proceedings references relieving rhodes robert ronald rosenfeld roth schapire sciences sense senseval siglex statistical stephen system systems taipei taiwan theoretic third tity toolkit ulrich using vincent weifeng with word workshop xavier yoram
http://acl.ldc.upenn.edu/P/P05/P05-2005.pdf	85	Exploiting Named Entity Taggers in a Second Language	abstraccao abstraction acontecimento adaboost adaptation advances alessandro algorithm analyzers andrew annotation annual answering anthony appear applied approach april architecture arevalo artifact artificial association astrof aurelio automatic automatically babych based bikel bogdan borthwick budapest building canada canaria cards carlos carreras catalan chapter christian chunk class classifier coisa computational conference conll constantine context control cooperation coutino coverage cucchiarelli daelemans daniel data date david development dictionaries disi document domain eacl edii editors edmonton eibe electronica enrich entities entity entropy espanola european event external extractor fabriani features february fifth finder fine flexible florian formal frank georgios gideon gomez gonzalez grained gran guodong hartley high house hungary iberamia implementations improvement improving indexing informatics information insti intelligence internal international issue january java jesus jian johnson jose karkaletsis kaufmann language learning learns lecture lema lenguaje lexical linguistics local location lopez lrec luis machine management mance mann manuel maria marquez mart mateo maximum method methods mexico michel mihalcea miles miller minimization mining miscellaneous missikoff moldovan montes montse morgan nacional name named nantzintla natural networks notes noun november nymble object obra ontologies ontology optica organizacao organization osborne other overall padro pages paliouras palmas paola paolo para perez perfor performance person pessoa petasis pineda portuguese practical press probabilistic proce proceedings processing programs proper proposal puebla quality quantity question quinlan rada radu ralph recognition references research resources retrieval reyes richard risk robust samiento schwartz scott semanet semantic september series sica sigir simon simple sociedad solorio spain spanish special springer spyropoulos stacking studies system systems table tagger tagging taipei taiwan techniques technology tempo text thamar that thesis through tong toni tools tors translation tributed tuto university using valor vangelis variado velardi villasenor walter weischedel what wide with witten workshop xavier york zhang zhou
http://acl.ldc.upenn.edu/P/P05/P05-1073.pdf	74	Joint Learning Improves Semantic Role Labeling	aarseth acknowledgements activity advanced andrew annotated answering aquaint arda argument authors automatic baker bank beatrice berkeley building calibrating capable carreras charles charniak christopher classification classifiers colingacl collin collins comments computational conclusions conditional conll corpus current cynthia daniel data dependencies development discriminative discussions distance ecml emnlp english entropy especially eugene extraction features fernando fields fillmore framenet frames gains generalized generative gildea hacioglu harabagiu have helpful icml incorporating inference information insightful inspired intelligence introduction intuition james john jointly journal jurafsky kadri kingsbury krugler labeling lafferty language large learning levy like line linguistic linguistics long lowe machine machines manning marcinkiewicz marcus martha martin mary maximum mccallum michael mihai mitchell model modeling models naacl natural nianwen over pages palmer parser parsing part paul penn pereira pradhan predicate probabilistic proceedings program project proposition punyakanok question random references reflecting reranking research reviewers roger role roles roth rquez sameer sanda santorini segmenting semantic sequence shallow shared shown structures substantial suggestions support supported surdeanu task thank that their there this thompson treebank true useful using valerie vasin vector verbs ward wayne when williams with work would xavier yuancheng zimak
http://acl.ldc.upenn.edu/P/P05/P05-1023.pdf	24	Data-Defined Kernels for Parse Reranking Derived from Probabilistic Models	algorithm algorithms american annual association barcelona brian broad budapest chapter charniak classification collins computational conf coverage discrete discriminative driven duffy efficient entropy eugene european freund head henderson history human hungary implementation incremental inducing inspired james joint kernels language large learning linguistics machine madisson margin maximum meeting michael model models natural nigel north over pages parser parsing pennsylvania perceptron philadelphia proc ranking references rens representations reranking roark robert schapire seattle spain stanford statistical structures tagging theory thesis university using voted washington with yoav
http://acl.ldc.upenn.edu/P/P05/P05-3019.pdf	127	SenseRelate::TargetWord ­ A Generalized Framework for Word Sense Disambiguation	academic acapulco adapted algorithm archive artificial association august automatic availability available banerjee barcelona boston carroll cicling city comprehensive computational concepts cone conference cpan cream daniel database demonstration development diana dictionaries dictionary disambiguation dist distributed dordrecht dumais editor editors eighteenth electronic extended fass february fellbaum finding fourth freely from gloss http ijcai intelligence intelligent international jason john julie july kluwer koeling lesk lexical lexicon license linguistics machine main marcu massachusetts mccarthy mcdonald measure measures measuring meeting mexico michelizzi naacl network open overlaps pages papers patwardhan pedersen perl pine plate platform predominant press proceedings processing providing public pustejovsky readable references relatedness roukos salim search semantic semantics sense senserelate senserelatetargetword senses siddharth sigdoc similarity slator source sourceforge spain susan targetword tell text third tools tractable under untagged using volume weeds wilks word wordnet written
http://acl.ldc.upenn.edu/P/P05/P05-3030.pdf	138	Organizing English Reading Materials for Vocabulary Learning	agree automating based basic beginning beishikku bijuaru chujo college conclusion constructing contain context corpus courseware data development difficult distribution document effect eigo english fostered found frequency from genung goiryoku have hitoshi industrial isahara isolating journal keenbow kouka learned learning level many masao material materials measuring method midori nihon nishigaki niyoru okapi paclic pages part preparation prepare presenting proc process proposed rather reading references robertson sample selecting sentei shisaku shokyuushamuke sofutowuea sono specialized students table tanimura teach teachers technology test than that them this toeic trec uchibori university ushida utilizing utiyama very visual vocabulary walker while words yamazaki yoosei
http://acl.ldc.upenn.edu/P/P05/P05-1043.pdf	44	Learning Stochastic OT Grammars: A Bayesian approach using Data Augmentation and Gibbs Sampling	acquisition addition algorithm algorithms also american amsterdam analysis angeles anttila appeal approaches arbisi association asudeh augmentation based bayesian becomes berlin boersma calculating calculation cambridge carlo change computing connection consider constraint constraints current data deletion densities difficult dimensions diminutive direction directions distribution distributions early editor editors empirical enough entropy example experimentation extended extract faetar final finnish first fixed fixing formation forthcoming found from future gelfand gelman geman generative gibbs goldwater gradual grammar grammars have hayes high identifiability ieee images inference inferences information informative inquiry institute intelligence interaction interesting introduce issue iterative johnson joint journal kager keller kelm language learn learning linguistic linguistics literature machine make marginal marginals master maximum mixture model modeling modes monte more moreover morphology motivated multiple nagy needed normal notice number often optimality optionality order other pages paper parameterized parameters pater pattern phonetic phonological phonology posterior press prince priorities priors probabilistic probability proceedings provide rankings references relaxation reliable represent restoration reynolds rubin sampling science sciences scientific second sequences series simulation since smith smolensky some sophisticated spanish spenader springer springerverlag stages stanford statistical statistics stochastic stockholm strategies tanner tests that their theories theory therefore thesis they this through trans treated types ucla university unknown using variability variable variances variation which with within wong word work workshop would zonneveld
http://acl.ldc.upenn.edu/P/P05/P05-1.pdf	0	ACL-05	abhishek accuracy acero acupressure again alex alfio algorithms alon alspector amigo amit amounts analysis ando andreas andrei andrew anna annual another anselmo appendix approximate approximation argument aria arul arun association author automatic ayurveda balanced banko bannard barbara barzilay based bilmes biology blood body bond boulis brian brill briscoe broad broder burberry burch calisthenics callison canada carberry cardio carlo carpuat categorization cavnar center chanel charikar charniak chelba chemical cherry chester chew chiang choosing chosen chowdhury chris christ christoph christopher church ciprian civil classification claudio clustering cohn colin coling collins complexity compression computer computing consider considered constantinos containment contents context coverage crammer curran curse dagan dang daniel dataproblem david davide deane deepak define dekai dekang demir detection detmar dickinson diego dimensionality ding dior disaster discovering document documents dong dubey earthquake economics edmonton eduard efficient eisner electrical elizabeth elzer engineering enrique environmental eruption estimation eugene eugenio evans fair feldenkrais felisa fendi fernandez fernando ferragamo filali final fingerprinting finkel fossati francis frank from fujita function functions gang geffet general gideon gildea ginzburg giuliano glass gliozzo goemans gonzalo gram green greenwood grenager grishman guan gucci guodong habash hacioglu haghighi haifeng hailstorm haizhou haller hanks harabagiu harper harvard have heaven henderson heng here hickl hideki hill hindle hiroya hong hovy hurricane hutchinson hwang ichi improved index indyk information ingrid integers introduction inui isozaki ivan ivona jacm james japan jason jeff jenine jenny jens jeremy jian joakim johanna john johnson jonathan josh julio jurafsky just kadri karim kate keith keller kenji kingdom klein koby koehn kohomban kolcz korhonen kristina kucerova kudo kulick kyoto landslide lapata lauren lavie lehmann lexicography lexicon liang lillian lists louis maayan macwhinney manabu mann manning mapping marcu mari marine mark markus martha martin mary math matsuzaki maximum mcdonald mcgill mcgraw means mechanical menezes meurers michael miles ming mirella mitigating miyao modern moens moldovan montreal moore moses motwani mouse mudslide muller murat mutual nancy near nearest neighbors never nice nilsson nivre nizar noah norms noun oepen okumura osborne ostendorf owen paiva palmer pang pantel parser patrick paucity paul penas people pereira permutation pete philadelphia philipp physics pilates pittsburgh polynomials possibilities possible power prada pradhan predicate principar principle problems proceedings program programming publications qigong quality quirk rabin radu ralph rambow random randomization raquel ravichandran references reflexology regina removing replica report reprographics research resemblance retrieval rieser roark robustness roger rose rounding ryan sagae salton sameer sample sanae sanda sandra saraclar sarah sasaki sathyajith satisfiability scaling schlangen schroeder schwarm science scott semidefinite seniz senses sequences seth shaw shiatsu shimei shouxun shriberg shubin sigkdd signature similar similarity smith sook soricut space spade stat stephan stephanie stevenson stoc stolcke strapparava structures such susan suzuki symposium table takaaki takamura takashi taku takuya tanaka techniques technology texas text theory therapeutic there third thus tidal tillmann titov tong touch toutanova towards trang trenkle trevor trnka trond tsujii tsunami turner typhoon unique university unlv upali using vancouver vegas verdejo verena vincent volcanic vuitton wang ward wave wayne weld white williamson windstorm winters with word words would xiaofeng xiii xviii yallop yang yarowsky ying young your yuan yusuke yutaka zhang zhanyi zhao zheng zhou zukerman
http://acl.ldc.upenn.edu/P/P05/P05-1029.pdf	30	Scaling up from Dialogue to Multilogue: some principles and benchmarks	aamas advances agent agents amount analysis approach association automatique autonomous barcelona barry based best cambridge carletta catalog clarification clark claws computational conference cooper corpus dabbs dialogue dialogues dignum dimensions discussion editors ellipsis engineering english experimental fabra fernandez fipa first foundation frank garrod garside gerard ginzburg group harlow herbert http information intelligent interaction interactive international invited issue james jean joint jonathan language languages linguistics longman management meeting monologue multi natural nicholas nonsentential pages paper party physical pompeu practice pragmatics press proceedings process protocol psychological psychology raquel references resolving robin roger ruback science semantics serial simon social special specifications spoken state structure study system systems tagging testbed theoretic toulouse towards traitement type university update using utterances vocal vreeswijk word workshop
http://acl.ldc.upenn.edu/P/P05/P05-1046.pdf	47	Unsupervised Learning of Field Segmentation Models for Information Extraction	able accurate advertisements applications approach aspect barzilay based bibliographic blei building catching chen cikm citation citations class classification classified clustering computational conclusions conditional conference constraining content demonstrated different documents domain domains drift engines entropy examined extraction field fields francisco freitag from fundamentals further generation hall have hearst hidden hierarchical hofmann icml identity ieee ijcai including information interest international into juang kaufmann knowledge learned learning linguistics machine management markov marthi matching maximum mccallum methods milch mixture model models moreno morgan multiparagraph naacl nigam nips pages papers passages pasula peng pereira popat prentice press probabilistic proceedings quality rabiner random recognition references refinements rennie research restrict russell search segmentation segmenting selected sets seymore showed shpitser sigir similar small space specific speech structure subtopic summarization supervised task tenth text texttiling that this those topic toutanova training tutorial uncertainty unsupervised used using vaithyanathan were with work yield
http://acl.ldc.upenn.edu/P/P05/P05-1021.pdf	22	Improving Pronoun Resolution Using Statistics-Based Semantic Compatibility Information	abney addition affect alternative american anahora anaphora anaphors annual appelt apply approach approaches argument association automatic back based bean believe bigrams bridging candidate cardie cascades centering chapter chunk chunking combination combinations common compatibility competition computational compute conclusion conference configuration contextual coreference corpora corpus could dagan deep dialogue different domain domains effective effectively empirical entity european evaluated even experiments exploration explored finite focussed framework francisco freqencies frequencies future have hitzeman hobbs hybrid improvement improving incorporated information international interpretation issues itai japan kaufmann kehler keller knowledge language lapata large learning limited lingua linguistics lisbon logic look machine majority markert maroudas meeting mehta methods mitkov model models modjeska more morgan most muller named natural neutral never newswire nissim north noun obtain obvious other overall pages parsing partial performance personal philadelphia phrases poesio portugal predicate proceedings processing programs pronominal pronoun pronouns proposed publishers quinlan recognition references research resolution resolve resolving results riloff robust role school semantic showed simma single soon source sources specifically spoken state statistics still strube such summer systematically tagger take taylor technical text that twin under unseen unsupervised using utility utilize where while with work workshop would yang zhou
http://acl.ldc.upenn.edu/P/P05/P05-1045.pdf	46	Incorporating Non-local Information into Information Extraction Systems by Gibbs Sampling	aaai abbeel abney analysis andrieu annealing applications approach artificial arts attribute bayesian biomedical biomedicine borthwick bunescu canada carnegie chieu choi clark coling collective collocation computational conditional conference conll connections context cowell curran data dawid della diego dingare discriminative distant distributions domains doucet eacl edmonton entities entity entropy expert exploiting extraction features fields finkel francisco freitag freitas from gazetteers gelatt geman gibbs global grammars grover hidden hmms icml ieee images independent inducing informal information intelligence introduction joint jordan kaufmann kirkpatrick koller laarhoven labeling lafferty language lauritzen learning leek linguistics machine malouf manning markov master maximum mccallum mcmc mellon mikheev models moens mooney morgan named natural networks nguyen nissim optimization other overcoming pages pattern pereira philip pietra probabilistic proceedings processing publishers rabiner random recognition references reidel relational relaxation restoration science segmentation segmenting selected sequence shrinkage simulated sparseness speech spiegelhalter springer statistical stochastic sutton syntax systems tagger taskar theory thesis transactions transitions tutorial uncertianty university using value vecchi verlag with without workshop york
http://acl.ldc.upenn.edu/P/P05/P05-2018.pdf	98	Centrality Measures in Text Mining: Prediction of Noun Phrases that Appear in Abstracts	analysis artificial based boston bowyer centering centrality chawla class combined communication complex computer conference corman define discursive document dooley erkan explicitly given hall human imbalance india information intelligence international introduction japkowicz jurafsky kegelmeyer knowledge kuhn language martin matrix mcphee michigan minority natural ncii organizational over path point problem proc processing radev references research resonance sampling significance smote speech stephenson strategies studying synthetic systems technique therefore understanding university value where write written zelen
http://acl.ldc.upenn.edu/P/P05/P05-1074.pdf	75	Paraphrasing with Bilingual Parallel Corpora	aligned alignment approach barzilay brown burch callison chris computational corpora corpus david della diab estimation extracting from june kathleen learning lillian linguistics machine mathematics mckeown mercer method miles mona multiplesequence naacl osborne parallel parameter paraphrase paraphrases peter philip pietra proceedings references regina resnik robert sense sentence statistical stephen tagging talbot translation unsupervised using vincent with word
http://acl.ldc.upenn.edu/P/P05/P05-1044.pdf	45	Contrastive Estimation: Training Log-Linear Models on Unlabeled Data	abney adaptive algorithm algorithmic altun annealing applications approach attribute based bootstrapping bound cald canon carnegie charniak clark classification college collins communication computational conditional conll constraint contrastive crammer curran data dempster disambiguation discriminative divergence early eisner emnlp english entropy error estimation estimators experts extraction feature fields finitestate forests from functions gcnu geman ghahramani grammar grammars grammatical guiding hinton hofmann icml icslp ieee ijcai implementation incomplete induction inference investigating isit jebara johnson joint journal juang katagiri kernel klein kuhn label labeling lafferty laird language large latent learning lexicalized likelihood limited linear linguistics logic london loss machine machines manning markov mathematical maximization maximum mccallum measures mellon memory merialdo method methods minimizing minimum miyao mmie model modeling models multiclass naacl namedentity natural nips nocedal odell optimization osborne parameter parsing path pentland pereira prescher press principle probabilistic problems proc processing products programming random ratnaparkhi recognition references report reranking research results riezler rivaling rosenfeld roukos royal rubin scale schuurmans segmenting semi sense sequence sequences shallow signal singer smith society speech statistical stochastic structure supervised systems taggers tagging tarjan technical techniques text thesis towards training trans transducers tsujii tubingen unification unified universitat university unlabelled unsupervised using valtchev value vector vocabulary wang ward with woodland word workshop yarowsky young zhao
http://acl.ldc.upenn.edu/P/P05/P05-1039.pdf	40	What to do when lexicalization fails: parsing German with suffix analysis and smoothing	american amit amsterdam annual applied assigning association automata basil beil bikel blackwell blaheta booth brants carroll center chapter charniak chen college collins computational computing conference daniel dependencies detlef driven dubey empirical england enriching entropy estimation eugene ewan formal frank franz function gazdar generalized geoffrey gerald german glenn goodman grammar harvard head ieee inside inspired intricacies ivan japan joshua keller klein language languages lexicalized linguistics maryland mats maximum meeting michael model modeling models naacl natural north outside oxford pages park parsed parser parsing part pcfg pennsylvania performance phase prescher probabilistic proceedings processing pullum references rens report representation research riezler rooth sapporo seattle sister smoothing speech stanley statistical statistics stefan structure study switching symposium tagger tags taylor technical techniques technology tenth text theory thesis thorsten university washington with
http://acl.ldc.upenn.edu/P/P05/P05-1067.pdf	68	Machine Translation Using Probabilistic Synchronous Dependency Insertion Grammars	advances aligned alignment alshawi annotation august automatic baltimore bangalore barcelona based bikel bilingual bleu bonnie boris brown center clsp cohesion coling collections collins comp companion comparison computational conference context corpora correspondence crammer curin daniel decoder decoding dekai della dependency description design different ding divergences dorr douglas driven eisner emnlp engine estimation evaluating evaluation extracting fast fernando final finite first formal formalism france franz from generation germann gildea graehl grammar grammars hajic head heidi hermann hopkins ibrahim ijcnlp insertion international inversion isomorphic jahr japan jason jimmy john johns joint jonathan josef july katz kenji kevin knight koby kolak lafferty language large learning lingual linguistics loosely machine mappings marcu margin mathematics mcdonald melamed mercer method michael model models monolingual multi multitext naacl natural okan onaizan online optimal pairs palmer papineni parallel parameter paraphrases paraphrasing parsers parsing pennsylvania pereira perspectives peter philadelphia philip phrasal pietra proceedings processing projection proposed purdy rebecca recent references report resnik robert roukos ryan sapporo schabes second shieber smith solution spain speech state statistical stephen stochastic structural summer synchronous syntax systematic technical thesis training transducers transduction translation translational tree treeadjoining treelet ulrich university using various vincent volume ward weinberg workshop yamada yarowsky
http://acl.ldc.upenn.edu/P/P05/P05-2025.pdf	105	Unsupervised Discrimination and Labeling of Ambiguous Names	acknowledgments address advisor also ambiguity amit amruta anagha annual association automatic automatically bagga baldwin barcelona based better bigram boston breck canada carroll city classes clustering clusters computational conclusions conference context contexts continual cross dataset david deepak depends diana dimensionality disambiguation discrimination document edmonton encouraging entity estimating experiments extended finding first gideon guenther guidance hastie have highly increases intelligent international john journal julie koeling kulkarni labeling language learning like limited linguistics localized mann mccarthy meeting mexico model montreal naacl name natural number order pages pantel patrick pedersen performance performs personal predominant problem proceedings processing purandare quebec ravichandran references referencing representation results robert royal schutze scope second selection semantic sense senses series shown shows significant similar similarity simple sixth society space spaces spain statistic statistics support technique techniques text thank that tibshirani trevor unsupervised untagged using vector walther weeds which with word work would yarowsky
http://acl.ldc.upenn.edu/P/P05/P05-2002.pdf	82	Understanding the thematic structure of the Qur'an: an exploratory multivariate approach	about above abundant academic addressing also amesterdam among analysis anderson applied approach arabic arnold august based been believers black brown cambridge chapman classification cluster clustering coling combinations computation computing conceptual concerned conclusion conduct constitute construction corpus data definition diego different dimensional directions discovering discussed distance distinction dror dunn each edinburgh everitt evidence figure flynn foundations frequency from future generated gordon gore group groups hair hall hand handbook have high however indicate indicative international interpretation interrelationships introduction issues jain kachigan language languages learning length lexical limitations linguistic linguistics literary london main mannila manning mass mathematical measure message mining modeling mohamed more morphological multivariate murty narratives natural neural oakes occurrences ongoing only people plot preliminary prentice press principles proceedings processing provide radius references relative resolved results review reward righteous script section semantic shaharabani smyth standardization statistical statistics stemming structure such supportive sura suras surveys switzerland talmon tatham thabet that their thematic themes these those tinsley tree trees trends university usable useful variables variation verleysen when wintner with work workshop york
http://acl.ldc.upenn.edu/P/P05/P05-1063.pdf	64	Discriminative Syntactic Language Modeling for Speech Recognition	academic acoustics ahmad alex algorithm algorithms american andreas andrej andrew annual anoop arxiv association based bocchieri brian brown bunt canada canon carroll chapter charniak chelba chen chuck ciprian college collins computation computational computer computes conditional conference connectionist constraint context conversational corrective daniel data david dependencies dependency developments discriminative distribution dordrecht down dragomir driven editors edmonton effectiveness efficient emami emnlp empirical enrico eric estimation estimators eugene experiments exploiting exponential features fernando fields fosler franz fraser frederick free from gary geman generation gildea giorgio goodman gram grammar grammars harper harry harvard head hidden hopkins http human icassp icml ieee immediate initial integrating integration international investigating izhak jain jelinek john johns johnson jonathan josef joshua jurafsky katherine kenji khudanpur kluwer knowledge kumar labeling lafferty language large libin linguistic linguistically linguistics ljolje machine mark markov mary mccallum meeting methods michael model modeling models morgan motivated multiple murat naacl natural nelson north pages parameter parsing peng pennsylvania perceptron pereira practice precise predictive prefix probabilistic probabilities probability proc proceedings processing publishers purdue radev random recognition references report rich richer riezler riley roark robust ronald rosenfeld sanjeev saraclar sarkar satta segal segmenting sentence sequence shafran shallow shankar shen signal smith smoothing smorgasbord sources speech stanley statistical stefan steven stochastic stolcke structure structured stuart study submitted substring superarv syntactic system tajchman technical techniques technology that theory thesis tightly training transcription translation unificationbased university using vehicle viren vocabulary wang whole williams williamstown with wooters workshop xiaojin yamada zhen zhiyi
http://acl.ldc.upenn.edu/P/P05/P05-1065.pdf	66	Reading Level Assessment Using Support Vector Machines and Statistical Language Models	accessed advances alignment approach augmenting barzilay bigrams books boulis branch brookline burges bylsma callan cambridge categorization chall charniak chelba chen chisson cjlin classification clear collins comparable computer conference conjunction corpora csie dale dalechall data derivation difficulty educating elhadad emnlp empirical english enlisted entropy european feature features fishburne formula formulas goodman guide gunning hill http inspired instruction ireland jelinek joachims kernel kincaid language large learners learning machine machines making malagon many mass maximum mcgraw memphis methods mining modeling monolingual naacl naval navy office olympia ostendorf pages papers parser personnel practical predicting press proc public readability reading redundancycompensated references relevant report representation research revisited rodgers scale scholkopf selection sentence siam smola smoothing speech state station structured study superintendent support technique techniques text thompson vector washington with words workshop writing york
http://acl.ldc.upenn.edu/P/P05/P05-1012.pdf	13	Online Large-Margin Training of Dependency Parsers	acquisition aggressive algorithm algorithmic algorithms analysis annotated applications articles automatic based bikel building censor charniak clark classify coling collins computational conditional corpus crammer crouch culotta curran czech data dekel dependency deterministic ding discriminative driven emnlp english entropy estimation experiments extraction fields from functional grammar grammars grishman guestrin hajic head hidden icml implementation incremental insertion inspired intricacies iwpt jmlr joachims johnson journal kaplan kernel kernels king klein kluwer koller labeling lafferty language large learning lexical linear linguistics machine machines manning marcinkiewicz marcus margin markov matsumoto maximum maxwell mccallum methods model models multiclass naacl natural networks news nips nivre online optimization oxford palmer parallel paraphrase parse parser parsing part passive penn pennsylvania perceptron pereira press probabilistic problems proc ramshaw random ratnaparkhi references relation riezler roark santorini scholz segmenting sekine sequence shalev shallow shinyama shwartz singer sorensen speech statistical street sudo support synchronous tagging taskar techniques text theory thesis tillmann training translation tree treebank ultraconservative university using vector wall with yamada zenios
http://acl.ldc.upenn.edu/P/P05/P05-2021.pdf	101	Speech Recognition of Czech - Inclusion of Rare Words Helps	adaptive annotation arpa backoff barbora based book broadcast byrne canada catalog categories coling computer conference consortium continuous corpus czech data dependency entropic entropy eurospeech evaluation finite finke geutner hajic hajicova highly hladka human icassp icring inflectional inflective international ircing issues jarmila jelinek jindrich josef khudanpur krbec language languages large linguistic lisbon ludek malach models mohri morphological mountreal muller multilingual news number pages pajas panevova pass pavel pereira petr portugal prague prediction proceedings project pruning psutka radova recognition references republic resources rich riley ruda scheytt seattle sgall speech spontaneous state stolcke structured tagging tagset technology tousek transcribing transducers treebank using vidova vlasta vocabulariesfor vocabulary washington weighted william workshop young zeleznaa
http://acl.ldc.upenn.edu/P/P05/P05-1014.pdf	15	The Distributional Inclusion Hypotheses and Lexical Entailment	academic association automatic barcelona bernardo challenge challenges chapter chklovski church coling computational contextual dagan dale dekker discovery distributional emnlp entailment exploration feature fine geffet geneva glickman grained grefenstette gregory handbook hanks harold harris hermann information kenneth kluwer language lexicography linguistics maayan magnini marcel mathematical mining moisl mutual natural norms oren pantel pascal patrick proc processing publishers quality recognizing references relations semantic similarity somers southampton spain structures switzerland textual thesaurus timothy vector verb verbocean wiley word workshop zelig
http://acl.ldc.upenn.edu/P/P05/P05-3007.pdf	115	High Throughput Modularized NLP System for Clinical Text	abbreviation acronym advancing agreement american amia annotated annotation annotations annual apiii appear applications approach aronson association automating based bigram biomedical biometrics categorical chapman chute clinical close coden combination combines complemen comprehensive computational corpus correction crowley data developing different disease domain effective entropy epidemiological evaluation fairly francisco friedman general gnegex highest however iinformatics implementation important incidence informatics innovation instruction international issue issues journal koch landis language lexicons linguistics manually mapping maximum mcinness measurement medical medinfo meeting metamap metathesaurus method methodology methods mitchell models more narrative natural negation network normalization notes observer pakhomov part particularly pathology patients pedersen performance philadelohia phrases practice precision prevalence print proceedings processing processor produced program queries radiology recall recruiting references reports research results score semi shared shows special specific speech spelling statistics studies supervised symposium system systems table tagger tagging text texts than that these this thompson through tools towards trials umls using valued washington where which
http://acl.ldc.upenn.edu/P/P05/P05-1019.pdf	20	Modelling the substitutability of discourse connectives	aaai abdessamad academic acquiring acquisition advaith alex alistair ambiguity american analysis andrew annual appear application approach artificial asher association august automatic based boston brigitte brunswick cambridge casimir center chapter charles choice cognitive cogsci coherence cohesion computational computer conference connectives contextual conversation correlates csli curran daniel data david determinants diab discourse discovery distributional driven echihabi edinburgh editor eduard effort emnlp empirical english environmental european evaluation explorations extraction fourth framework general generation george goodenough grammar grefenstette gregory grote halliday hasan hobbs hovy human hutchinson improvements inferring information intelligence international interpretation james japan jerry jersey journal julie july kaufmann kehler kluwer knott kulikowski language lapata lascarides learn lexical lillian linguistics lisbon logics longman lrec manfred marcu marker markers mateo mcdonald meaning measure measures measuring meeting methodology methods miller mining mirella modelling moens mona moore morgan moser motivating natural nicholas ninth north pages parsing philadelphia philip planning portugal practice preserving press problems proceedings processes processing publications publishers recognizing reference references relations report research resnik resources rubenstein sapporo science scott second semantic sentence sentenceinternal sholom siddharthan similarity simplifying society spring stanford stede structure study summarization symposium synonymy systems taxonomy technical technology temporal text that theory thesaurus thesis twenty university unsupervised usage using verb weeds weir weiss when william workshop
http://acl.ldc.upenn.edu/P/P05/P05-1033.pdf	34	A Hierarchical Phrase-Based Model for Statistical Machine Translation	americas appear applied assembler association based beam berlin bikel block brown byrne california center chen chiang chinese computational computer computing conference daniel david decoder della deng directed editor emnlp empirical engineering estimation evaluation example finite foundations franz goodman hans harvard incremental interpretation josef joshua journal koehn kumar language linguistics machine marcu mathematics mercer methods model modeling models naacl natural noun pages parameter parsing peter pharaoh philipp phrase pietra proceedings processing pushdown references report research robert sciences search second shankar significance sixth smoothing southern speech springer stanley state statistical stephen study synchronous syntax system technical techniques technology template tests thesis transducer translation translations treebank ullman ulrich university verbmobil verlag vincent wahlster weighted william wolfgang workshop yonggang
http://acl.ldc.upenn.edu/P/P05/P05-1059.pdf	60	Stochastic Lexicalized Inversion Transduction Grammar for Alignment	acknowledgments adam aligning alignment alshawi american andrew annual apparatus assistance association august bangalore based berger better bilingual both bracketing brown chapter chinese christopher coling collections comparative computational conclusion conference constraints context corpora corpus daniel dekai della dependency description designed douglas edmonton english estimation exact experiments fast fillett finite formal france franz geneva gildea grammar grammars grateful head helped hermann hiyan hong improved international inversion japan josef kehler kenji kevin klein knight kong language learning lexicalization lexicalized limitations linguistics machine manning mathematics meeting melamed mercer method model models multitext naacl north october pages parallel parameter parse parsers parsing patent peter pietra presented procedure proceedings proposed pruning reasons rebecca references reordering result richard robert sapporo selection sentences shona showed sides smoothing specially srinivas state states statistical stephen stochastic study supervised switzerland synchronous syntax tanslation techniques than that toulouse training transducers transduction translation united unlexicalized unsupervised using very vincent viterbi with words yamada zens zhang
http://acl.ldc.upenn.edu/P/P05/P05-1038.pdf	39	Lexicalization in Crosslinguistic Probabilistic Parsing: The Case of French	abeille accurate alexandra ambiguous american amit analysis anne annotated annotation annual applied artificial association athens barcelona based beatrice bikel boston brants brigitte broad building cambridge chapter charniak chiang chinese christoph christopher clement college collins computational conference context corpus coverage czech daniel david decision deep dekai dekang dependencies dependency design distributional dubey editors efficient emmanuel empirical engine english entropy eugene european evaluating evaluation francisco frank free french from generative geneva german giguet grammars hajic hans harder head helmut highly hong human inspired intelligence international jacques joint kaufmann keller kinyon klein kong krenn lance language languages large levy lexicalised lexicalized lingual linguistics lionel madrid magerman manning marcinkiewicz marcus mary maryland maximum meeting memory method methods michael mitchell model models montreal morgan multi natural north ofspeech order pages parallelprocessing park parse parser parsers parsing part penn probabilistic proceedings processing ramshaw references research resources roger santorini sapporo scheme schiehlen schmid seattle sister skut statistical strategies syntactic tagging technologies technology thorsten three tillmann tree treebank university unlexicalized using uszkoreit vectors vergne washington with wojciech word workshop
http://acl.ldc.upenn.edu/P/P05/P05-1047.pdf	48	A Semantic Approach to IE Pattern Induction	anniversary applications architecture association based bontcheva cambridge cardie circus computational conference conrath corpus cunningham database description development editor electronic fellbaum fisher fourth francisco gate international jiang lehnert lexical linguistics massachusetts maynard mccarthy meeting message pages philadelphia press proceedings references research riloff robust semantic similarity soderland some statistics system tablan taiwan taxonomy understanding university used wordnet
http://acl.ldc.upenn.edu/P/P05/P05-3014.pdf	122	SenseLearner: Word Sense Disambiguation for All Words in Unrestricted Text	about active addition algorithm also analysis antwerp applied approach attempts august automatic average back barcelona based baseline bayes bosch both bunker chooses class coling communication comparable competitive complexity computational conclusion concordance conference content created daelemans darpa data database decadt default defined described difficult directions directly disambiguate disambiguation disambiguatuion each efficient evaluated evaluating evaluation examples exceed experiment experiments expert explained faruque feature found france frequent from future gambl general genetic guide hendrickx higher hoste however human improve improves individual instance international into jersey july language large larger leacock learner learning lexical likely linguistics memory method mihalcea miller minimally model models moldovan more most naive note number obtained open optimization over pages paper part pattern performance philadelphia plainsboro precision proceedings randee recent reference references rely report resulted results seem selection selects semantic sense senselearner senses senseval separate sets siglex significantly simple sloot some spain speech successes supervised surprisingly system systems taipei taiwan technical technology test text that these third this thus tilburg timbl toulouse training university unrestricted using verbs version very were when where which while with word wordnet words workshop yuret zavrel
http://acl.ldc.upenn.edu/P/P05/P05-2001.pdf	81	Hybrid Methods for POS Guessing of Chinese Unknown Words	adwait andy applied based basic beijing bing brants category chao chen china chinese chooi coling combining computational conference contemporary corpus detection duan emnlp enhanced entropy guessing hong huiming identification institute international japan jiang jiann journal language learning ling linguistics master maximum method methods ming models morphology nara natural nlprs pages part partof peking proceedings processing ratnaparkhi references report richard rule science shiwen speech sproat statistical statistically system tagger technical technology thesis thorsten tutorial university unknown word words workshop xuefeng zixin
http://acl.ldc.upenn.edu/P/P05/P05-2004.pdf	84	Jointly Labeling Multiple Sequences: A Factorial HMM Approach	accuracies accuracy achieved acknowledges acknowledgments acoustics adding advanced also annotated anonymous applied applying approach artificial augment augmenting author automatic bach backoff bartels based basenp bayesian between bigram bilmes brants brill buchholz building case changing chapter chris chunker chunkers chunking comments competitive computational conditional conf conference conll construction constructive context corpora corpus cyclic data decision dependency directions discriminative discussions driven dynamic dynamically emnlp endto english entropy eric error especially examine factored factorial feature features fernando fhmm fields figure finally first florian forests foundations francis frederick from function further future gang geiger generalized generative ghahramani gmtk good grace grant graphical gratefully heckerman hidden huang icml identification improve includes inference insightful integrate intelligence interactions intl introduction jeff jelinek joint jordan katrin kirchhoff klein knowledge kudo labeling language large learn learning like linguistics machine machines manning mappings marcinkiewicz marcus markov matsumoto maximum mccallum measure merit michael model modeling models multi multidimensional multinets multiple naacl naccl natural network networks netwrosk ngai number ofspeech open overall parallel parsing part peng penn pereira pitch plan planned press problem proc proceedings processing promising radu ramshaw random ratnaparkhi recently references representation research results reviewers rich rohanimanesh sang santorini schutze second sentences sequences series shallow shared signal similar similarity singer sizes smoothing software source speech state statistical study such support sutton switch switching system tagger taggers tagging task techniques terms text thank that their third this thorsten three time tjong toolkit toutanova tracking training transformation treebank trees trigram type uncertainty under unified using varying vector very which will with word words workshop would zhou zweig
http://acl.ldc.upenn.edu/P/P05/P05-1018.pdf	19	Modeling Local Coherence: An Entity-based Approach	academic acquisition across analysis annotated annual applications approaches asher automatic barzilay based birmingham bleu building cache cambridge cardie catching centering clarendon classification clickthrough cluk coherence cohesion collins colloquium computational computed computer content conversation coreference corpus current dale data discourse drift dumais eacl emnlp engines eugenio evaluating evaluation experiments expert exploring foltz framework from functional generation generative getting gram grosz grounding hahn hasler hirst hitzeman hovy hpsg ieee improving indicator induction information instantiations into investigation joachims joshi karamanis kaufmann kernels kibble kintsch knowledge kuhn kukich kulikowski landauer language lapata lascarides latent leaf learn learning lexical lexicalised linguistics local logics machine mann manning marcu markova mellish message method methods metrics miltsakaki model modeling models morgan mori morris naacl natural nets neural oberlander occurrence optimising optimizing ordering pami papineni parametric parse parsing path plato poesio power practice prediction press prince probabilistic problem proceedings proceesings processes projection psychological recognition references referential reiter relations reliably representation research resolution review rhetorical role rough roukos scott search selection semantic sentence shift skills solution souza speech statistical statistics stevenson string strube structure structuring summaries summarisation summarization systems teaching text textual that theory thesaural thomson three toutanova transactions transitions translation trees university using view walker ward weinstein weiss with writing zock
http://acl.ldc.upenn.edu/P/P05/P05-2024.pdf	104	Corpus-Oriented Development of Japanese HPSG Parsers	adam akira amano approach berger bond bresnan chen chikara cmucs computational della entropy eric francis fujita gaussian grammars grammatical hashimoto hinoki ijcnlp introduction kaname kaplan kasahara language linguistics maximum mental models nariyama natural nichols ohtani pietra press prior proc processing references relations report representation representations rosenfeld sanae shigeaki shigeko smoothing stephen takaaki tanaka technical text treebank understanding vincent
http://acl.ldc.upenn.edu/P/P05/P05-2006.pdf	86	Automatic Discovery of Intentions in Text and its Application to Question Answering	achieve alqaeda anscombe arpa asian attempting attitudes audi base based bell bratman bruce bunker cambridge claudia communication complex computational concordance cooperation cornell could data database destruction does east from george harvard html http human increase increasing influence info ingredient ingredients intend intending intention intentions ithaca janyce journal known korea laden language leacock learning lexical linguistics manner martha martin mass massachusetts matthew means melanie mental michael military miller mining network north osama philosophical philosophy plans pollack possibility practical press proceedings purchase purpose putin qaeda quinlan randee reason reasoning rebecca references region remains report restore review robert ross rulequest russia said semantic subjective technology tengi that their theresa tools treaty tried trying university weapons what where wiebe wilson with wordnet workshop york
http://acl.ldc.upenn.edu/P/P05/P05-1060.pdf	61	Multi-Field Information Extraction and Cross-Document Fusion	aaai acquisition adaptive agichtein answering answers approach approaches aseltine automatically banko barzilay bases biographical brill brin bunescu cardie carroll case clarke collections collective colognet combining computational concepcion conceptual conditional conference constraint constraints context cormack corpus cowie crystal czuba dalmas data database dataintensive dictionary diego document dossier dumais dynamic ecml edbt elhadad elsnet empirical examples exploiting extending extracting extraction factoid factorize fields fisher frame freitag from fusion generating gravano hidden hmms hovy huffman icdl icml ijcai inducing information international knowledge korelsky kushmerick labeling lafferty language large learning leek lehnert line linguistic linguistics lynam machine mallet mani markov master masterson mccallum mckeown mining model models molina mooney multi multilingual multiple nahm natural networks nirenburg notes pages patterns pereira personal perspectives pierce plain prager probabilistic proceedings processing producing profiles question questions radev random ravichandran redundancy references relational relations riloff rohanimanesh rosenfeld salgado satisfaction schiffman schmelzenbach segmenting sequence seymore shrinkage sigir snowball soderland sources spring statistics structure summaries summarization surface sutton symposium system technology text texts theoretical thesis threads toolkit trec untagged using wagstaff webber webdb white wide with working workshop world wvlc
http://acl.ldc.upenn.edu/P/P05/P05-2010.pdf	90	Using Readers to Identify Lexical Cohesive Structures in Texts	abelson accepted across adrian affect agreement alan allow although amorrortu analysis anaphora anchored annotated annotating annotation annotators answering assessing assigned association assumes attitude automating average averaging barbara barbu barcelona barzilay based beatrice behavioral better bonny boston building burger byron calculation cambridge carletta carthy castellan catalina categories category chaining chains chanahan classical classification classified classify cmplg coders cohesion cohesive coling colloquium communications comprehension computational computing conclusion conjunction constantin constructing content core coreference corpus corpusbased corr current daarc daniel data database david definite delaware description developing discourse distributions dodrecht donna editor editors electronic elhadad employs english enterprise equation erlb estibaliz estimate eugenio evans expected experiment experimentation experiments fellbaum finding first flood follows glass goals graeme group halimi halliday happens hasan hearst hill hillsdale hirschman hirst however human identification indexing inquiry intelligent inter international into investigation item items james jane janyce jean john jones journal july justified kappa kazman klaus knowledge krippendorff large lawrence levels lexical lexicography line linguistics lisa longman look lynette magdalena marc marcinkiewicz marcu marcus marti mary massimo mcgraw measures michael miller mitchell mitkov moldovan morris much multi naacl netherlands news nicola nonparametric novischi number often orasan pages paper paragraph passages patricia patterns penn phenomenon plans poesio pragmatics presented press proceedings processed psycholinguistic publications question raters reach reader reading reem references regina related relations reliability reliably renata resolution resources richard rick robert robinson roger role romera rumelhart ruqaiya ruslan sage santorini scalable schank sciences scripts second seed segmentation segmenting select semantic semantics showed sidney siegel similar smeaton sotirova spain springer standards statistic statistical statistics stokes story strategies structures studies subjectivity subsequent subtopic summarization system tagging tasks temporal text texts texttiling than that this thresholds through tools training treebank trees typical understanding used uses using validation various version very vieira vilain violeta ways webber wiebe with wordnet work workshop
http://acl.ldc.upenn.edu/P/P05/P05-1022.pdf	23	Coarse-to-fine n-best parsing and MaxEnt discriminative reranking	about accepted accuracy actual agreement american annotater annual argonne artificial association automaton bank benson bikel bilexical bound california chapter charniak collins computational computer conference context curfman currently daniel discriminative doubt efficient eisner entropy error estimate eugene european figures francisco free generative giorgio grammars head icml implementation improvement inspired instead institute intelligence inter international intricacies jason jorge kaufmann laboratory language learning lexicalised linguistics linguists lois machine manual massachusetts maximum mcinnes meeting michael model models morgan national natural north pages parser parsing penn probably proceedings reasonable reduction references rens report reranking revision sarich satta science seventeenth short stanford statistical steve submission take talking technical technology terry there this three tree underestimates upper users well
http://acl.ldc.upenn.edu/P/P05/P05-1005.pdf	6	Learning Semantic Classes for Word Sense Disambiguation	algorithm analysis anne antal bart based beze bosch cambridge classifier combination computational conference conrath context corpus cotton crestan daelemans daelmans database decadt disambiguation edmonds electronic english evaluating evaluation fellbaum france gambl genetic guide hoste improving international intl jakub jiang kool learner level lexical linguistics loupy measure memory monitored multi optimization overview press proc proceeding proceedings reference references report research second semantic sense senseval similarity sloot statistics systems task taxonomy technical text third tilburg timbl toulouse veronique version view walter with word wordnet words workshop zavrel
http://acl.ldc.upenn.edu/P/P05/P05-2026.pdf	106	A Domain-Specific Statistical Surface Realizer	adjoining adwait also annotated automatically bangalore based being bracketed bratt brill case cavedon chen cheng collecting computational computer corpora corpus correctness coverage data dialogs domain domains driven efficiently empirical eric error express extraction failed fernando framework from full future geary general generated generation generator given grammars harriman human icslp inlg inside irene island jeju korea langkilde language learning linguistics means methods mishra model naacl natural navigation newark niekrasz offers only other outside overgenerating paris parse parses part partially pereira peters proc processing purpose query rambow ratnaparkhi reestimation references schabes search seattle semantic sentence sentences shriberg sometimes speech spoken srinivas statistically step structure study suited surface tagging talana that this trainable trained transformation tree upson using verification well weng wizard work workshop
http://acl.ldc.upenn.edu/P/P05/P05-3029.pdf	137	HAHAcronym: A Computational Humor System	aaai about acapulco acronyms applied approaches artificial attardo berlin beziehung binsted boston cognition cognitive communication computational computer conference creating deutike development dordrecht editors effects enschede february freud gabora getting gruyter hofstadter hulstijn human humancomputer humor humorous humour ijcai implemented intelligence interaction interface international issue joint jokes kernal laboratory lancaster leipzig linguistic logic measurement mechanisms mediated memo mexico minsky model morkes mouton nass national netherlands nijholt oriented password proc proceedings punning raskin references report riddles ritchie ruch seattle seine semantic sense serious special stock strapparava swordfish synopsis task technical theory twente twlt unbewussten unconscious university verbal vienna witz workshop
http://acl.ldc.upenn.edu/P/P05/P05-1050.pdf	51	Domain Kernels for Word Sense Disambiguation	algorithm american analysis athens barcelona based bosh cambridge cavaglia codes computer cristianini daelemens dagan decadt deerwester disambiguation domains dumais exploitation field furnas gambl genetic gliozzo greece harshman hoste indexing information integrating into introduction journal july june landauer language latent lexical lrec machines magnini memory optimization pages press proc proceedings references science semantic senseval shawe society speech strapparava subject supervised support taylor university unsupervised vector wordnet
http://acl.ldc.upenn.edu/P/P05/P05-1049.pdf	50	Word Sense Disambiguation Using Label Propagation Based Semi-Supervised Learning	belkin blum brown cald categories chapelle classification cluster coling corpora data disambiguation fields from functions gaussian ghahramani harmonic icml kernels label labeled lafferty large learning manifold mercer methods mincuts models nips niyogi partially propagation randomized reddy references report robert roget rwebangira scholkopf semi semisupervised sense statistical stephen structure supervised tech trained unlabeled using vincent weston with word yarowsky
http://acl.ldc.upenn.edu/P/P05/P05-1028.pdf	29	Exploring and Exploiting the Limited Utility of Captions in Recognizing Intention in Information Graphics	about access advances aimed along also alternative analysis annoyance appear appears attention autobrief automated automatic average based bayesian briefings cambridge caption captions carberry carenini charts chester clark class classes communicate communicative complex composite computer computers conceptual conf conference consider constraint contain contains contributes coreferences corio corpus current currently data default demir demonstrate detection dhillon diagrams dialog differently difficult discourse djia document documents does durfee editors effectiveness efficient effort elzer european evidence examine example experimental extend extracting first focused framework from frustration fully futrelle future generate generation getting giuseppe given goals graphic graphics graphs green grice grouped half handle happen have hoffman huber human hunter hypothesizes icslp ijcai impairments incorporating individuals industrial information integrated intelligent intended intention intentions interfaces international into investigate itself johanna joint jones journal kerpedjiev kinds knowledge krupski label labels language lapalme larger limit line lnai mani mapping mattis maybury meaning message methodologies methodology moore more multiple nancy natural need network nikolakis occasionally occurrence occurs other ourselves paper parsing part particular patterns perceptual philosophical plan plans present presented press primary probabilistic problem proc proceedings processing project prosody provided providing rare reason recognising recognition recognizing references reiter research resolving results review roth secondary series shallow showed shriberg sight signals some spoken sripada stephan steven stolcke strategy studies study such summarization summary symposium system systems task tasks text texts than that third this time turbine uncertainty understanding university used user users uses using utilizing utterer verb visual wellman what when will with work workshop zukerman
http://acl.ldc.upenn.edu/P/P05/P05-1027.pdf	28	Question Answering as Question-Biased Term Extraction: A New Approach toward Multilingual QA	about academic action adam aircraft airport amusement analysis animal appendix approach award berger book broadcast city color company computational conference country countx crime currency date della disaster disease drug entropy ethnic evaluation event extent facility firstranked form frequency from games geological government group information inst institute landform language linguistics location maximum measurement method military mineral money month monument movie multiplication museum music name nationality natural newspaper numex object offense organization paragraph park party percent period periodx person phenomena phone phrase picture pietra point port position printing processing product province public qbte question railroad rank references region religion results road rule school seismic ship show space spaceship speed sports station stephen team theory these time timex title train training type types used vegetable vincent volume water weapon week weight worship year
http://acl.ldc.upenn.edu/P/P05/P05-1036.pdf	37	Supervised and Unsupervised Learning for Sentence Compression	aaai achieved acknowledgements added adding additional allow also always american angheluta annotated annual approach artificial assigning association audio automatically barbara based beatrice better between beyond blaheta blind bloedorn blur bonnie building chapter charniak clarification collins come comments compress compressed compression compressions computational computationl conclusion conference constrain constraints corpus could created cure daniel data david deletion design desired discover distinction document dorr easiest enforcing english eric eugene example extraction find forest francine francisco function further gates generation generative goals good grammaticality grant grefenstette gregory have head hoped however immediate improved improvements improves improving inderjeet intelligence intelligent irene jing johnson kaufm kevin knight labels langkilde language large larger learned leuven lexicalised like limit linguistics lower mani marcinkiewicz marcu marcus marie mark mary meeting michael michell mitra model models moens more morgan much national noisychannel north notes original other over pages paradigms parameter paraphrase parsed parsing particular penn perform performs possible probabilistic probably problems proceedings producing provide quantities rates reduction references relate remarkably revising richard richer roxana rudradeb rule rules santorini scanning schwartz semisupervised sentence sentences service show simple some spring statistical statistics step success such summaries summarization supervised supported surely symposium syntactic system tags task telegraphic test text than thank that their them there this three topiary training treebank type understanding unsupervised using utility version weighting well which with work working would xiuli zajic
http://acl.ldc.upenn.edu/P/P05/P05-1075.pdf	76	A Nonparametric Method for Extraction of Candidate Phrasal Terms	absent accurate acquisition addison adopting algorithm ananiadou annual anything applications applied approaches array arrays artificial assesses association associations automatic baayen balancing based behavior being best better between bigrams boguraev boston cambridge candidate carroll case catching chap characterization chengqing cheshire choueka church cognition coincidence coling collocation collocations combined combining comparaitve comparison computational compute conclusion conference congress connection constraints content cooccurrences corpora corpusbased cvalue dagan daille data databases description dias dictionary differs digital dispersion distribution distributions domain dordrecht dunning each efficient effort empirical engineering euralex evaluation evert expansion expectation expressions extension extracting extraction extractor fair fakotakis family ferreira first foundations fourth frantzi frequency friends from garcia general german gold good gram grams guillor hanks haystack headwords higher houghton human identification identifying implementation implications including independent indexing indicates induction information institut intelligence interesting international interpreted jacquemin johansson journal jurafsky justeson kanguage katz kennedy klavans kluwer knowledge knowledgefree kokkinkais krenn kuwer language large learning least lecture length lexical lexicography lexicon libraries likelihood linguistic linguistically linguistics lingusitics list local locating looking lopes lrec manning maschinelle masks mass mathematics matwin maxima maynard measure measures measuring meeting method methods metric metrics mifflin mima ming mining model morphology multidimensional multiword mutual nagata natural needles ninth noncompositional nonparametric normalization norms notes number occurrences order outperform over pages pairs pantel pereira philosophy phrasal phrase phraseological phrases portion positional press principle problem proceeding proceedings processing properties proposed psychobiology qualitative quality rank ranks ratio realization recognition references related resnik results retrieval retrieving returning riao rigid schone score seems sekine selectional semantic series shimohata sicilia silva sixth smadja smith solved some sprachverarbeitung springer squared standard statistical statistics strenght strength stroulia structures study stuttgart suffix suggest sugio surprise symbolic synergy syntax taln technical techniques technology term termight terminology terms text textual than thanapoulos that their theoretic thesis this thus translating tsujii tzoukermann understated unit units university unrestricted upon using value variant varying verb verlag wesley word words workshop xtract york zipf zipfian zong
http://acl.ldc.upenn.edu/P/P05/P05-3027.pdf	135	SenseClusters: Unsupervised Clustering and Labeling of Similar Contexts	actually airlines ambiguous american among approach asked assigned august automatic automatically avoid best bill bold boston both bruce bush case charles city classifier clustering clusters cognitive computational conference contexts contextual correlates cruise currently descriptive develop discovered discriminating discrimination distinguishing email empirical equal exists experimental experiments face fact february finally gates general george have improve intelligent international kulkarni labels language learning linguistics majority manually methods mexico miller most must name names natural note number optimal pages pedersen proceedings processes processing provided providence purandare references results roles schutze second semantic sense senses serve setting shows significantly similar similarity sixth spaces stop successfully suggests table taken text that these this those uncategorized untagged upon user value vector well which will word working
http://acl.ldc.upenn.edu/P/P05/P05-1007.pdf	8	Aggregation improves learning: experiments in natural language generation for intelligent tutoring systems	aaai aggregating aggregation aied alternatives analysis anaphora annual anyway applied approximate argument argumentative armin artificial association barbara based beno bhembe bodvarsson boyle brazil brighton bringing bruce brunin building caldwell cambridge canada carberry carenini case cessation chicago chinese ching chris clause coch coding cognitive coherence coling commonsense comparing computational computer concise conciseness conference content conversation corpus dale dalianis davide department descriptions developing diag diagnostic dialogue dialogues douglas dynamic editors education effectiveness ehud empirical empirically eugenio european evaluating evens exemplars experiments explanation extensible failure fast fiedler fifteenth fifth forbes formal fossati framework from generating generation generators giuseppe glass gnome graesser grice haller harvey heena helmut hercules holding hong horacek huang human iconoclast illinois influence information inlg institute instruction integrating intelligence intelligent international iwanska james jeon johanna jose journal just kibble knight knowledge kong lake language lavoie learning lessons lester letters liesl linguistics litman logic long maarika maceio manual maxim mcdaniel meeting mellish michael mike moore natural niagara ninth nominal operators ordering osman owen pages paraphrasing person plan plans portable porter power practical practitioners press proceedings processing publishing pytlikzillig quantity rambow raval realizer reape reasoning references regularities reiter report representation research researchers richard riley robert robertson robust rodger roma rose rovick sandra science select seventh shapiro shaw silliman sixteenth smoking sneps society specifications spitkovsky spoken stocholm strong structure studies study susan synthesizing systems tailored technical techniques technology terrence text textproduction thesis third three together towne traat trolio tutorial tutoring typed understanding university using vanlehn versus what while white with workshop xiaoron young
http://acl.ldc.upenn.edu/P/P05/P05-1054.pdf	55	A Quantitative Analysis of Lexical Differences Between Genders in Telephone Conversations	acoustics american argamon author automatic automatically backing based between blackwell cambridge categorization categorizing cieri classification clustering coates communication computational computing conf conference conversational corpus differences doddington dude eckert editor editors empirical european eurospeech evaluation extensible extensive fakotakis feature fisher forman gender generations genre ginet gram http icassp icslp idiolectal improved international intl kiesling kneser kokkinakis koppel language learning lexical linguistic linguistics literary lrec machine mccallum mcconnell measures metrics miller modeling next pages pilot press proc proceedings processing publishers reader recognition references research resource resources retrieval richness selection shimoni signal singh speaker speakers speech speechto spoken stamatatos statistical stolcke study technology terms text texts toolkit university walker written
http://acl.ldc.upenn.edu/P/P05/P05-1026.pdf	27	Experiments with Interactive Question-Answering	aaai aarseth abductive acquired acquisition anaphora answer answering answers approach argument association automated automatic automatically based bases believe bensley bowden chin claimed clark coling collection combining complete completely computational conclusions conference considered consistently counter coverage created cybernetics databases depends describe describes discovery distance does domain domanin dudani eacl easy eduard environment evaluation expanded experiments expository extraction factor factors faqfinder found from gave general geneva good grishman harabagiu hearst helped helpful hovy human huttunen ideal ieee incremental information infratructure insights interactions issues john kameyama knowledge language like line linguistics links lytinen make marti match meaningful meeting megumi methods mihai mining model modeling moldovan multi narayanan nearest neighbour novel open operational overall pages papers paragraph pasi patterns paul perspective practical predicate presented proceedings promoted provided quabs quality question questions ralph ready reasoning recognizing references referential representations resolution respondents results retrieval robust roman rule sanda satisfaction satisfied scale scenario search segmentation semantic signatures silja specific speed spring srini stimulated structures study suggestions summarization surdeanu survey switzerland symposium system systems table tapanainen techniques technology text texts that their then they thinking this three tomuro topic topics training transactions trec twelfth types understanding unrestricted used useful user users using weighted were williams with work workshop would yangarber
http://acl.ldc.upenn.edu/P/P05/P05-3021.pdf	129	Automating Temporal Annotation with TARSQI	about academic allen andrea andrew annotation applications arosio aspect assertive association automatic beth bierwisch building carol chanod chapter clause collection communications complex computational conference corpus david dependency discourse dordrecht eacl editor editors elra erich evaluation extraction fabrizio fact factivity ferro finitestate first france from gaizauskas graham granada hanks hans heidolph hooper information international intervals james jean ject joan john kamp karl karttunen katz kimball kiparsky kluwer knowledge language lauri lazo leffa lexical linguistics lisa logic madrid maintaining manfred marcia mokhtar mouton natural netherlands object observations pages papers paris patrick paul pierre predicates press proceedings processing progress publishers pustejovsky references resources reyle robert roser salah saur semantic semantics sentences setzer some spain spatial subi sundheim syntax temporal tense timebank toulouse transducers using vilson volume workshop york
http://acl.ldc.upenn.edu/P/P05/P05-3010.pdf	118	Learning Source-Target Surface Patterns for Web-based Terminology Translation	academic algorithm amta base bilingual chang chinet coling collocations communication computational corpora data dictionary evaluation extraction feature from graehl grams hutchins introduction journal knight koehn kwok linguistics machine monolingual nagata name noun parallel personal phrase phrases press proc references rich saito smadja somers statistical suzuki system translation transliteration using workshop xtract
http://acl.ldc.upenn.edu/P/P05/P05-2016.pdf	96	Dependency-Based Statistical Machine Translation	abeille academic adam alena alignment american analysis anne annotated annotating annotation annual approach april argument arpa assignment association automatic barbora based beatrice berger bies bleu bohmova britta brown building bulletin canada canaria candide categories channel chapter charniak clsp cmejrek cocke cohesion coling computational conference corpora corpus curin czech daniel david decoding dekai della dependency dzeroski eacl editor edmonton emnlp empirical english entropy estimation eugene evaluation fast ferguson ferty final fourth france franz franzjosef fredrick functor germann gillett grace gram grammatical gran hajic hajicova harry havelka head heidi hermann hladka hongsing human immediate improved inflective inspired international issues jahr jelinek john josef july june karen katz kenji kevin kishore kluwer knight koehn lafferty language languages large learning level linguistics lrec lubos machine macintyre marcinkiewicz marcu marcus mark martin mary mathematical mathematics maximum meeting melamed mercer method methodological methods michael mitchell model models montreal morphological mtfinal natural ngram nist noah north occurrence onaizan optimal pages palmas papineni parameter parser parsing paul penn peter petr petra philip phrasal phrase pietra prague predicate prediction printz procedures proceedings processing projects publishers purdy quality references report resources rich robert roossin roukos salim santorini saso scenario schasberger sgall smith spain speech statistical statistics stephen stochastic structure structured study syntactically syntax synthesis system tagging tagset technical technology tectogrammatical tests theoretical third three todd toulouse transfer translation treebank treebanks ulrich ures using vincent volume ward weijing with wong workshop yamada yarowsky yaser zabokrtsky zdenek
http://acl.ldc.upenn.edu/P/P05/P05-1041.pdf	42	High Precision Treebanking -- Blazing Useful Trees Using POS Information --	abeille abney academic adam ageno agreement akira alicia amano analyses analysis anne annotation annotator annotators antonia applications asian athens bender beyond black bond borja brant brants bufi building carl carter cast chapter chicago chikara christoper civit claudia coling comparing completing computation computational computers conference construction copestake corpora corpus correction coverage cycle daniel database david deep detecting detmar development dickinson donald driven editor efficient emily engineering english environments eric estimation evaluation evolution ezra familiarity first flickinger formalisms forthcoming framework francis fred from fujita gdaniec german grafting grammar greece grishman grove hainan hans harrison hashimoto head hindle hinoki hiroshi holistic http humanities ijcnlp improving inconsistencies ingria inter international into introduction ipsj island issue ivan japanese jelinek joint joseph journal judith kaname kanasugi kasahara kaufmann kilgariff klavans kluwer kondo kristina kurohashi language lanl lexeed lexicon lieberman lingo linguistic linguistics lrec madrid maintenance makoto manning maria mark markus mart melanie mental meurers minimal modeling montserrat morgan motivation nagao nariyama natural navarro newspaper nichols nuria oepen ohtani pacific pages parsed parsing philip phrase pollard preliminary press procedure proceedings processing publishers qualitative quantitative quantitatively ralph recursion redwoods references research resources results revision robert rosenzweig sadao sanae sato sean second semantic semantics senseval shallow shieber shigeaki shigeko siegel size skut special speech spoken standardization statistical stephan steven structure strzalkowski stuart supervised swee sweeden syntactic system tadahisa taipei taiwan takaaki tanaka testing text theories thorsten tokyo tomek tomoko tool toutanova towards training treebank treebanker treebanks tsujii understanding university using uszkoreit vaxjo volume wallis while with wojciech word workshop
http://acl.ldc.upenn.edu/P/P05/P05-2017.pdf	97	Minimalist Parsing of Subjects Displaced from Embedded Clauses in Free Word Order Languages	able adjacency allow also analysis antje appear asad barco binding bkessler blackwell brett christian clauses commanding computation conclusions constituency constituent constituents constrained csli cyclicity demonstrated deutscher developing discontinuity discontinuous discourse displaced displacement distance dordrecht edition editor editors edward elements embedded escape espana estal even extracted features fine form framework free from future government grammar grammars guage haegeman hans have heed highly http iliano incremental inspired introduction islands italian jose kamp kessler kluwer languages latin left liliane linguistic long luigi luis maintain making mart maximi means minimalist move munoz natural order oxford pages parse parser patricio periphery position processing prosodic publications rafael reasons recognition references required requirement rizzi rohrer saiz sayeed sentence sentences springer stabler stan stanford stipulation structure subject system szpakowicz that theory this though thus took tsci using verlag vicedo were with word work wustl
http://acl.ldc.upenn.edu/P/P05/P05-1008.pdf	9	Empirically-based Control of Natural Language Generation	across adjoining analysis annual applied approach approximate architecture associates association barbara biber boosting brighton cahill cambridge carroll categories centering china cognitive coherence computational computer conference constraints control controlled corpus correctness coverage crude daniel david decision declarative deemter dialogue dimarco dimension dimensional discourse douglas eacl edinburgh edition eduard empirical engineering english erlbaum european evans exploiting factor fifth flexible focus forest formality framework frameworks from geary general generate generating generation generator grammar grammars green grosz hong hovy index inlg institute international investigating ircs irene issue italy john joshi journal july kathleen kong lancaster langkilde language lawrence leaflets lexicalized linear linguistics local lynne making marilyn mckeown meeting modelling multi natural nicolas nicolov nonhierarchical number output paiva parameters pennsylvania pharmaceutical pisa planner potential power pragmatic press proceeding proceedings purpose rags rambow references regression related representations research results revisited riches robert rogati sanford science scott sentence shaumyan sigley smets sons sorter sorting special speech spoken stephen stick strategies student style stylistic stylistically system text them thesis training tree under university using variation verification volume walker weinstein weir weisberg where wide wiley workshop writing written
http://acl.ldc.upenn.edu/P/P05/P05-3017.pdf	125	Supporting Annotation Layers for Natural Language Processing	advances alfonso analysis andrade andreas annotation architecture australasian automatic biological bird blaschke building cassidy catherine cawg chen christian christos communication critical database david davidson extending extraction formal framework from grishman grosse haejoong harrington information interactions isard ismb jonathan klein language level liberman linguistic management marion mark mate mckelvie mengel michael miguel moeller morten multi ouzounis pages planx proceedings processing program protein queries querying ralph references requirements saga scientific speech steve steven support survey susan system technology text tipster treebanks updating valencia workb workshop xpath yifeng zheng
http://acl.ldc.upenn.edu/P/P05/P05-2015.pdf	95	Learning Strategies for Open-Domain Natural Language Question Answering	action altun anlp answering approach artif based braz brill burger case charniak class classification comprehension computational cumby database deep driven electronic elements engineering error even facilitate fellbaum formal framework from gaizauskas garrett here hirschman ijcai intel kaebling khardon kosmala language languageprocessing learning lexical light linear linguistics littman machine moore moscovich naacl natarajan natural overview pang part prediction press problems processing programs question read reading references reinforcement relational representations research riloff roth rule salvo solutions speech speedup statistical study survey system tadepalli tagging take tests that thelen threshold track transformation trec using valiant view voorhees word wordnet yang zeller zohar zorn
http://acl.ldc.upenn.edu/P/P05/P05-3006.pdf	114	Descriptive Question Answering in Encyclopedia	answer answering approach association bilotti biographical blair caused chua combining comparative computational concepcion conference corpus corrected definitional depends descriptive details eacl earthquake ellen european felshin from further goldensohn heterogeneous hybrid influence information initial inverted katz large linguistic linguistics mani mckeown moreover multiple often overview patterns points poor precision principle problem proceedings producing question questions quiz realized recall references resources results retreival retrieval retrieved schlaikjer score sentence shiffman should sigir statistics study style such summaries system text that thirteenth this topic track trec twelfth twelve voorhees wave week what which workshop xian
http://acl.ldc.upenn.edu/P/P05/P05-3001.pdf	109	An Information-State Approach to Collaborative Reference	about activity adopting agreement another applying approaches assumptions because bleam both bridges cambridge clarification clark coextensive cognition cognitive collaborating collaboration collaborative collagen communicative comp computational computer computing content context conversational coordination core dale declarative devault dialogue directly discourse discussion distractor doran editors enables engine expressions flairs focus from gain generation gibbs gricean have heeman hirst human implications indeed information intelligence intention intentions interaction interface interpretation interpretations involves joint knowledge language larsson lesh ling linguistic linguistics london magazine management many maxims microplanning model modules move natural only pages palmer part people pragmatics process processes purver reaching reconciling reference references referring reiter relevant representations requests requires research rich science sets sidner situated social stack state stone structure studying substantive suggests suspect system systems tanenhaus that their theory there thesis toolkit traditions traum trindi trueswell univ using ways webber wilkes with work world
http://acl.ldc.upenn.edu/P/P05/P05-1068.pdf	69	Context-dependent SMT Model using Bilingual Verb-Noun Collocation	alignment asia bilingual brown cambridge chunk clarkson complexity computation computational conference decoding della discussion esca estimation eurospeech extraction hwang information kevin knight knowledge kyonghee language linguistics machine mathematics mercer modeling models pacific paclic paik parameter peter pietra proc references rosenfeld sasaki sook squibs statistical stephen tokyo toolkit translation using vincent wordreplacement young yutaka
http://acl.ldc.upenn.edu/P/P05/P05-1032.pdf	33	Scaling Phrase-Based Statistical Machine Translation to Larger Corpora and Longer Phrases	algorithms amta annual arrays based beam brown computational corpus daniel decoder della dicrete draft estimation europarl evaluation first franz gene josef june koehn line linguistics machine manber marcu mathematics mercer method models multilingual myers naacl pages parameter peter pharaoh philipp phrase pietra proceedings references robert search searches siam statistical stephen string suffix symposium translation unpublished vincent
http://acl.ldc.upenn.edu/P/P05/P05-3004.pdf	112	CL Research's Knowledge Management System	about addition algol allow also amia analysis annual answering approach automatic available based biomedical buckland centering communications compiler components comprehensive computational conference constructed corpus daniel dedicated dictionaries dictionary directed document documents domain dynamically each egra eighth eleventh evaluation evolving experiments exploration explore extending extensive extensively first fiszman focused from full functionalities functions gaithersburg gged gildea harman have html http hypernymic informatics integrated interpreter izat joel jurafsky kilicoglu labeling learning linguistics litkowski machine main makes many medical more navigli networks nguage nist nlpir novelty objective oceedi ominent other parser parsing part particular patterns preceding prepositions processor projects pronoun proposition publication pubs question readable references relation research resolution retrieval rhetorical rindflesch roget role roles sections semant semantic sites special specially strategies style subcategorization summar summarization summary supplemented surface symposium syntax system task tenth text texts that then thesaurus thesauruses these this trec triples unified used user using velardi verbs voorhees warehouses which whole with wordnet
http://acl.ldc.upenn.edu/P/P05/P05-1031.pdf	32	Towards Finding and Fixing Fragments: Using ML to Identify Non-Sentential Utterances and their Antecedents in Multi-Party Dialogue	aaai acknowledge acknowledgements adam agreement anonymous antal approach artificial assessing august available baldridge based berger bosch bunt carletta christophe classification classifying cohen coling comments computational computing conference corpus daelemans della dialog dialogue discourse discussions downloads during early editors effective elham ellipsis entropy fast fernandez florida fragment from garofolo geneva ginzburg gregory group guide helpful howard http induction intelligence interest jakub jason jean john jokinen jonathan july kappa kluwer knowledge kristiina language lappin laprun learner learning like linguistic linguistics machine martial maximum mcroy meaning memory michael michel muskens national natural nits orlando pages papers philadelphia pietra proceedings processing project raquel reference references report resolution reviewers rule sentential shalom shards sigdial simple singer sixteenth sloot special stages stanford statistic stephen strube study susan switzerland tabassi tasks technical third tilburg timbl university utterances version vincent volume walter william with workshop would yoram zavrel
http://acl.ldc.upenn.edu/P/P05/P05-1051.pdf	52	Improving Name Tagging by Reference Resolution and Relation Detection	absolute acknowledgements active additional advanced agency alexandria algorithm algorithms also among analysis andrew aone applications applied apply applying approach barcelona based baseline because best between bikel binary boosting borthwick capture carpuat character chieu chinatsu chinese chow christian coling collins combine combining compared components computational computer conf coreference cost could cross daniel darpa data decomain defense dekai demonstrated dept described detection diego dissertation dmitry document does earlier effectiveness efficiency efficient entity entropy error evaluation event example expect exploiting explore explored extend extended extra extracting extraction features fifth finder finding foundation framework from fung further gain global good government grant grishman heng here hidden hierarchy highperformance hwee hypotheses identification improve improvement improvements including incorporate individual information integer intelligent interaction interactions involved jason kambhatla keeping language languages lattice learning leong levels lexical linguistics lists lufeng marine markov maximum measure meeting mention methods michael miller models multiple naacl name named namedentity national natural necessarily need nominal nymble obtaining opportunity organize other overt paper papers particular pascale perceptron performance policy position possible presented proc procedure processing programming projects provided quite ralph ranking rate recognition reduce reduction reference references reflect relation relations required research resolution resources results richard rules scheffer schwartz science scott segmentation segmentations semantic sentence september short should some spain spawar specific speech stages stefan straightforwardly subtype such supported symposium syntactic system tagging taipei taiwan task techniques that these this tibbets tobias training under university using voted washington weischedel will with without word workshop wrobel york zelenko zhai
http://acl.ldc.upenn.edu/P/P05/P05-1016.pdf	17	Inducing Ontological Co-occurrence Vectors	agirre ansa applications baker barcelona berkeley canada chklovski church coling computers concepts corpus customizations disambiguating emnlp enriching extensions fillmore fine framenet gale grained hovy humanities large lexical lowe martinez method mining montreal naacl other pantel pittsburgh proceedings project references relations resources semantic senses signatures spain topic verb verbocean with word wordnet workshop yarowsky
http://acl.ldc.upenn.edu/P/P05/P05-3013.pdf	121	Language Independent Extractive Summarization	account algorithm algorithms also anatomy annual applied association authoritative automatic barcelona based because between brin bringing builds canada classification companion computational computer concept conclusion conference connections context corpus document drawn eacl edmonton emnlp empirical engine entire entities environment evaluation extraction extractive from gram graph graphs grimmett hirao hovy http human hyperlinked hypertextual identifies implements information intelligent into intuitively isdn isozaki iterative journal kleinberg language large lingusitics local madrid maeda meeting methods mihalcea moens naacl natural networks nilc nist nlpir occurrence only order oxford page pages pardo press probability proceedings processes processing projects random ranking recursively references rely report rino sasaki scalable scale search sentence sources spain statistics stirzaker summaries summarization system systems take tarau task technical technology temario teufel text textrank texts they through understanding unit university using various vertex volume well work workshop
http://acl.ldc.upenn.edu/P/P05/P05-2.pdf	79	ACL-05	adjoining adria adwait alex also anagha angus annotated asad automatically bangalore based beata beatrice beigman being bleecker bracketed bratt brill cakici case cavedon chen cheng collecting computational computer contents corpora corpus correctness coverage data dialogs diderichsen domain domains driven edward efficiently elming empirical eric error eugene express extraction failed fernando framework from full future geary general generated generation generator gispert given grammars grois harriman heidi huenerfauth human icslp index inge inlg inside introduction irene island ivanovic jakob jeffrey jeju jonathon kazuhiro kevin klebanov korea kulkarni langkilde language learning linguistics machek marta matt means methods mishra model naacl naglaa natural navigation newark niekrasz oana offers only organizers other outside overgenerating ozdowska paris parse parses part partially pavel pecina pereira peters petr philip podvesky postolache proc processing program purpose query rambow ratnaparkhi read reestimation references roberts ruken russell sayeed schabes search seattle semantic sentence sentences shriberg solorio sometimes speech spoken srinivas statistically step structure study suited surface sylwia table tagging talana tatu thabet thamar that this trainable trained transformation tree upson using verification viii well weng wizard work workshop xiaofei yoshida zhuli
http://acl.ldc.upenn.edu/P/P05/P05-3031.pdf	139	Reformatting Web Documents via Header Trees	algorithm algorithms analysis approaches automatic based categorization chang chen chia christina chung crescenzi cues data dempster discovery documents elements engineering evaluation extraction from gertz giansalvatore goodman guizhen hongjiang html icdar icde iepad incomplete information inrt joshua journal laird large likelihood manabu maximum mecca merialdo metrics michael mukherjee nanno neel okumura page pages paolo parsing pattern proceedings ramakrishnan references repetition reverse roadrunner royal rubin saikat saito semantic series shao sites society statistical structures structuring suguru sundaresan text tomoyuki towards valter visual vldb wenfang yang yiming yudong zhang
http://acl.ldc.upenn.edu/P/P05/P05-1040.pdf	41	Detecting Errors in Discontinuous Structural Annotation	abeille academic accuracy achieving adding almerindo almost analyses analysis anette anlp anne annotated annotation anomaly archer arthur automatic bartels beatrice beijing bergen berlin between beyond bies blache bracketing brants brigitte budapest building bulgaria bunt cacm canaria carroll ceedings centre china class coling combination comparison computation computational computing constituency constituent constituents continuous corpora corpus correct correction daelemans daniel dawn dekang detecting detection detlef detmar dickinson diego dipper discontinuous eacl edward effect eleazar english entity erhard errors eskin evaluation ferguson foundations frank fredkin free geoffrey george german gran gruyter guidelines halteren hans hansen harry heidelberg heike hinrichs hirst horck huck humanities hungary icslp improved improving inconsis inconsistencies ivan jakub japanese johansson john julia karel karen katz kawata kingsbury kopecek kordoni krenn kveton labelling language languages large learning level lezius linc linguistics lisbon lluis lrec luxembourg machine macintyre manual manually marcinkiewicz marcus mark markus marquez martha mcenery measures memory meurers mitch mouton muller multi naacl named noise norwegian ojeda oliva order padro pala palmas palmer parseval parsing part paul pavel penn pennsylvania petr philippe piao portugal prescher press proceedings processing project rayson recognition references research robert sabine santorini scheme scott seattle semantic silvia skut smith sojka sozopol speech spoken springer stefan stefanie stig style sweden syntactic system systems tagged taggers tagging tasks telljohann tency testing text their thorsten through tiger tony towards translation treebank treebanks trie tubingen ucrel university users uszkoreit valia vaxjo verbmobil wahlster walter washington with wojciech wolfgang word workshop yasuhiro york zavrel
http://acl.ldc.upenn.edu/P/P05/P05-3025.pdf	133	Interactively Exploring a Machine Translation Model	algorithm alignment approach bannard based broad burch cairo callison chris colin computational copestake coverage daniel decoder dekai development eamt editing english environment flickinger franz galley grammar hermann hopkins hpsg improved jahr josh kenji kevin knight linguistics lrec machine marcu mark michael michel model naaclhlt noah open polynomial proc references rule schroeder smith source statistical syntax syntaxbased template through time tool translation using visualization what workshop yamada
http://acl.ldc.upenn.edu/P/P05/P05-3026.pdf	134	Multi-Engine Machine Translation Guided by Explicit Word Matching	able algorithm also americas amta anlp annual applied approach architecture asru association automatic bangalore based basis berlin best better bleu bordel carbonell chooses choosing cken coling compared computational computing conclusions conference consensus consistent current data dataset eamt effectively engine error european evaluation even features final first fiscus focus font fourth frederking from further furthermore future generated generator germany good heads hogan hypotheses hypothesis identifying ieee improved improvement indicates italy jayaraman language languages lavie learning levin limited linguistics llitjos machine malta meeting memt meteor method metrics multi multiple names natural nirenburg obscured oracle original output papineni peterson philadelphia post privacy probst proceedings processing produces protect quality rates recall recognition recognizer reduced reduction references resources results riccardi roukos rover saarbr sagae score scores scoring select selects sentence sentences shown shows significance significantly similar speech springerverlag ssner stuttgart support system systems table than that their third this three thus tides tidhar trainable transfer translation understanding used valletta vogel voting ward washington will with word work workshop yield
http://acl.ldc.upenn.edu/P/P05/P05-1072.pdf	73	Semantic Role Labeling Using Different Syntactic Views	american arguments association bankruptcy barlow bartholomew bremmer brunk building charniak chen data dean deep emnlp entropy eugene features foster inference inspired japan john journal labeling linguistics maximum mining model naacl order owen pages parser predictive proceedings rambow recognition references restrictions robert sapporo seattle selection semantic statistical stine under variable washington wiley york
http://acl.ldc.upenn.edu/P/P05/P05-1053.pdf	54	Exploring Various Knowledge in Relation Extraction	aaai advances agichtein annual antonio aone applied april association barcelona becker brin california cambridge categorization chemnitz chichester cikm classification coling collections collins combining computational conference covention covolution cruz culotta cumby database dependency dietterich digital discrete dissertation driven duffy ecml edbt editors entities entropy european extending extract extracting extraction fawcett features from germany ghahramani gravano haussler head icml information international joachims journal july june kambhatla kaufmann kernel kernels knowledge language large learning lexical lexicography libraries linguistics machine machines management many march mateo maximum meeting message methods miller mishra models morgan natural neural novel online parsing patterns pennsylvania plain press probabilistic proceedings processing ramshaw reasoning recognition references relation relations relevant report research richardella roth santa seattle semantic snowball sorensen spain statistical structures supervised support syntactic systems taiwan technical technology text theory tree understanding university valencia vapnik vector washington webdb weekly weischedel whiley wide with wordnet workshop world zelenko zhang
http://acl.ldc.upenn.edu/P/P05/P05-3028.pdf	136	A Flexible Stand-Off Data Model with Query Language for Multi-Level Annotation	again allen also always annotation arcs association athens available behavior between bird both buneman carletta cassidy chiew child christoph come communication computers conference connecting contains corpora data database dialogue difference discourse distance either element elements evaluation evert explicit express flexible formal framework from graphs greece harrington heid hierarchical holger identify implicitly instruments intelligent international involved james japan jean jonathan journal july june katrin kilgour language level levels liberman like linguistic lisbon made management mark markable markables means methods michael milde mmax mmaxql modal model muller multi must nite only operator pado parent peter pointer points portugal proceedings queried queries query querying references related relation relations represented research resources result robertson same sapporo scale search sebastian section sequential shared sigdial similar simplified speech stefan steve steven strube supports system systems terms that thus time timealigned toolkit torsten towards ulrich ulrike using various voormann wang which with work workshop
http://acl.ldc.upenn.edu/P/P05/P05-1009.pdf	10	Towards Developing Generation Algorithms for Text-to-Text Applications	addison algorithms amalgam approach artificial association automata automatic bangalore barzilay bateman bleu brown california charles christian clifford cliffs columbia complexity computation computational computer conference cormen corston cucs decoding della department edition elhadad englewood eric estimation evaluation fernando finite formalism foundation functional fusion gamon geary generalpurpose generation giorgio graehl hall hatzivassiloglou hill hopcroft idlexpressions information intelligence international introduction irene jeffrey jersey john jonathan journal july kevin kishore knight langkilde language languages learned leiserson level linguistic linguistics london machine manual many mark mathematics matthiessen mcgraw mehryar mercer method michael model models modern module mohri moore multidocument natural nederhof norvig oliver overview owen pages papineni parameter paraphrasing parsing path pereira peter philadelphia pietra pinter prentice press probabilistic proceedings processing publishers rambow realization recognition references regina report representing research riley ringger rivest robert ronald roukos russell salim satta science second sentence simon southern speech srinivas state statistical stein stephen stuart summarization systemic technical text theory thesis thomas todd transducers translation transliteration tree ullman university user using vasileios version vincent ward weighted weijing wesley wordreplacement
http://acl.ldc.upenn.edu/P/P05/P05-1030.pdf	31	Implications for Generating Clarification Requests in Task-oriented Dialogues	aaai alexander bennett british burnard cambridge carnegie christina clarification clark communicator computing conference cooper corpus dialogue ellipsis gabsdil generation ginzburg guide herbert icslp international jonathan language malte mellon national natural oxford press proceedings processing reference references report resolving robin rudnicky services spoken spring symposium systems technical universiry university users using workshop written
http://acl.ldc.upenn.edu/P/P05/P05-2013.pdf	93	Automatic Induction of a CCG Grammar for Turkish	abeille addisonwesley ades adjective adjunct ahin ajdukiewicz akici analysis anne annotation arithmetic atalay baldridge beryl bilge bozs budapest building cambridge categorial category cattype collins combinatory comma computational conj control corpora data dedi derivation description determiner dilek dordrecht driven drop eacl east edinburgh editor figure free freq frequency frequent gaifman gibi given gokhan grammar grammars hakkani head hillel hockenmaier hoffman hungary icon information interface interpretation interpreted intransitive jason julia kazimierz kemal kluwer konnexitat language lexically lexicon lexicons like linguistically linguistics logic mark massachusetts master mccall michael middle models morph morphemic most nart natural noun occuring oflazer order oxford pages parsed parsing pennsylvania philosophy phrase polish postp press proceedings process quasi rank references ruken said sentence sentential shamir specified statistical statistics steedman storrs structure subject syntactic syntaktische syntax technical that thesis this token transitive treebank treebanks turkish type types university using verb which with word words workshop yehoshua zeynep
http://acl.ldc.upenn.edu/P/P05/P05-1070.pdf	71	Instance-based Sentence Boundary Determination by Optimization for Natural Language Generation	acapulco aggarwal aggregation algorithm also amanda american another anthony appeared approach architecture artificial authoring automated automatic background balancing barcelona based boosting boundary brazil brazilian brockenhurst canada capabilities capability capable case chen chris clear coherence coherent cohesion coling complex complexity comprehensive computational computer conclusion consensus constraints content conversation coordination cross data davey demonstrate dept desirable determination dialog dialogue discourse domain dynamic edinburgh ehud ellipsis english evaluation example failures first fortaleza generation generator graeme graphic gunning hill historical hybrid icnlg ijcai including independent information inlg instance instancebased intelligence intelligent interfaces jacques james john journal kennebunkport language lexical linguistics look maine maintaining mann many marilyn mcgraw mellish mexico michelle monica montreal moore multimedia multiparagraph naacl natural novel optimization ottawa over owen paper pittsburgh planner planning plausible possesses prasad presentation press preventing proc production properties proposed proteus providing psycholinguistically quality rambow rashmi rational realization reconstruction references reiter related report representative results revision ritchie robert robin rogati santa science sebastian segregatory segue selection semantic sentence severe shaw shimei sketches solution spain speech spoken stanford stent summaries superiority surface symposium syntactic system systems technique term text that this trainable training uist university using varges vikram walker waterloo wilkinson william work writing zhou
http://acl.ldc.upenn.edu/P/P05/P05-3012.pdf	120	Multimodal Generation in the COMIC Dialogue System	adaptation andrea appropriate arguments backoff baker bickmore billinghurst bilmes build campbell carenini cassell catizone chang clark comic contextually conversational dialogue domains eacl embodiment evaluating evaluative factored festival general generating giuseppe hannes intelligent interaction interfaces intonation isca jeff justine katrin kenny king kirchhoff korin language limited management mark michael models multimodal parallelized pittsburgh proceedings program project purpose rachel references richmond robert roberta setzer simon speech styles synthesis synthesizing systems thesis timothy university vilhjalmsson white wilks workshop yorick your
http://acl.ldc.upenn.edu/P/P05/P05-1011.pdf	12	Probabilistic disambiguation models for wide-coverage HPSG parsing	abney accuracy active artificial attribute auxiliary baldridge based canon carnegie charniak chen clark cmucs cohen coling collins computational conll coverage curran deep distributions driven dynamic empirical entropy estimation estimators exploiting gaussian geman grammars head hpsg importance inspired intelligence johnson kaplan king language learning linear linguistics maximum maxwell mellon methods models naacl natural osborne pages parse parser parsing pennsylvania press prior proc programming references report riezler rosenfeld selection shallow smoothing speed statistical stochastic supertagging technical thesis unification univ university using value vasserman wide
http://acl.ldc.upenn.edu/P/P05/P05-3020.pdf	128	A Practical Solution to the Problem of Automatic Part-of-Speech Induction from Text	alexander also analysis appropriate automatic barcelona based brown budapest capable clark classbased clustering coling columbus combining companion computational computed conclusion conclusions dayne della dendrogram desouza distributional each eacl figure found freitag from geneva gives global good gram hinrich improving indicate induction information insight inspired jennifer language linguistics local lower mercer models more morphological natural ofspeech paper part parts patterns peter pietra possible practical previous problem proceedings rapp references reinhard results robert scratch sense significantly similarities situa solution some speech study success summary syntactic table tagging text than that this toward unsupervised upper vectors vincent volume whereas wholecorpus with without word work
http://acl.ldc.upenn.edu/P/P05/P05-2007.pdf	87	American Sign Language Generation: Multimodal NLG with Multiple Linguistic Channels	access accessibility accessible achievement addition agents also american animation annual applications architecture association bahan baltimore being benefit benefits beyond boston cambridge capabilities cassell categories channel channels chapter characteristics churchill classifier college communication complex comprehension computational computer conference conventional conversational coordinated could critique data deaf demographic demographics design determined edition embodied encode english especially even exciting fact frequency from functional gallaudet generated generation gesture gestures grad grammar hard have hearing hierarchical holt http huenerfauth human iconic indirectly information instance institute integrated interaction interested interface interfaces internally international into intonation issues june kegl kopp language languages liddell like linguistic linguistically linguistics macfarlane machine maclaughlin many meaning meeting mentioned methodological microplanning mitchell models more morford multi multichannel multimodal multiple naacl natural neidle north only other output paper path pennsylvania people planning precise predicates press prevost proceedings produce producing prof progs prosody reading references report representations require requirement requires research researchers results richer school science serves sign signal signals spatial speech spoken stanford state states string structure student students studies subgroup sullivan survey syntax systems technical technologies technology tepper test text than that theoretical there these they this throughout timing topologically towards translation unique united universal university user users variety vegas vehicle vocal volume wish with words workshop
http://acl.ldc.upenn.edu/P/P05/P05-1020.pdf	21	Machine Learning for Coreference Resolution: From Local Classification to Global Ranking	aberdeen acquisition algorithm algorithms anaphora aone application applications applying approach approaches automated bagga baldwin bansal based bean bell bennett berger blum bunescu burger cardie chawla clickthrough clustering cohen coling collins combining competitive computational conditional conference connolly contextual coreference coreferencing correlation crossdocument cues data decision della dialogue discriminative distance driven eacl edit effective emnlp engines entity entropy error evaluating experiments extraction fast focs harabagiu hidden hirschman icml identity iida ijcai improving incorporating induction influence information integration inui ittycheriah jing joachims kambhatla kaufmann kehler knowledge language learning lehnert linguistics ller machine maiorano manual markov matsumoto maximum mccallum mccarthy mention message methods minimum mining model models morgan morton naacl natural noun optimizing pages perceptron phrase phrases pietra proach probabilistic proc processing programs pronoun proper pruning quinlan rapp reference references resolution riloff role roukos rule rules sample scheme scoring search selection sixth soon space spoken strategies strube synchronous takamura text theoretic theory tibbetts toward trainable training treatment tree trees uncertainty understanding unsupervised using vector vilain wagstaff wellner with workshop yang zelenko zhou
http://acl.ldc.upenn.edu/P/P05/P05-1066.pdf	67	Clause Restructuring for Statistical Machine Translation	about achieve action addressed alignment alshawi also amendments annotator anything approach area aspect automata automatic automatically avoid based baseline berger bilingual bleu bootstrap britain british brown category charniak chosen coling commission computational concrete conference confidence considerations consumer corpora current decisions dependencies discriminative dubey during edition editors efron emnlp endangering enlargement entropy error european evaluation examples extremely facilitating fact fall favour feature features figure figures food foreign found fraser from galley german gildea give graehl grammars hand hauliers have head here hinrichs hltnaacl hopkins however human hypotheses important improved improving included inconsiderable increase indeed information intervals into introduction inversion invited ireland issue issues jain jeopardise joint journeys judged keller khudanpur kingdom knight koehn kraftverkehrsunternehmen kumar language learned lehmann linguistics loosely lorries machine made main make marcu mathematics maximum mccord measuring melamed mentioned mercer method methodological metrics minimal minimum mission model models morpho must naacl natural necessity nice niessen nineties nothing noun number observations office order organisations originated other output pages panic paniksituation papineni parallel parsing particular patterns phrase phrases pietra practical presented priority probability problem proceedings processing produced proposal prosecutor public quoted radev random rate react recall recommendations reference references refers relates reordered reordering report representations reranking resign resources results rewrite rich roth roukos rule rushed sarkar scarce second seek sense several shen shipments side significance sisterhead smith smorgasbord social solutions some speak specifically springer statistical statistics stochastic subject summit syntactic syntax system take taken talk talking tenth testing tests than that their theoretical there therefore these they think this through tibshirani tiling tillmann training transducers transduction translation translations transport transporters travel travelling tree united using verlag veterinary vogel vote ward wasserman were what when where which whole whose will with without wong worse yamada yourselves zhang
http://acl.ldc.upenn.edu/P/P05/P05-1076.pdf	77	Automatic Acquisition of Adjectival Subcategorization from Corpora	accurate acquisition annotation annual applied association automatic beyond boguraev brent briscoe british burnard canaria canary carroll carter computational conference consortium contemporary copestake corpora corpus derivation dictionary empirical english evaluation extraction frames from general glenn graham grammatical grammaticallyindexed gran granada grover guide head high induction international islands jonathan language lexicalized lexicon linguistics longman mats meeting methods michael national natural oxford pages palmas parser parseval pcfg precision proc proceedings processing proposal reference references relational relations resources robust rooth sanfilippo schemes spain stanford statistical subcategorization survey taipei taiwan text third untagged users valence washington with workshop
http://acl.ldc.upenn.edu/P/P05/P05-1013.pdf	14	Pseudo-Projective Dependency Parsing	account acquired approximations automatically based bosch brill broad burke cahill campbell case categories charniak coling collins combination combining computational constituents constraint constraints coverage covington czech daelemans daum debusmann deep defeasible dependency dienes discontinuous distance donovan driven dubey duchier eisner empty entropy exploration foth genabith german grammar guide hajic head inspired konvens krbec kveton language learner linear linguistic linguistics long maximum memory menzel methods models naacl natural oliva parser parsing pcfg pennsylvania petkevic precedence principles probabilistic proceedings processing ramshaw recover reference references report resolution rules serial shallow sloot statistical statistics study syntactic tagging technical thesis three tilburg tillmann timbl topological trees university using version wide zavrel
http://acl.ldc.upenn.edu/P/P05/P05-1001.pdf	2	A High-Performance Semi-Supervised Learning Method for Text Chunking	analysis ando approach avrim blum carreras chieu colt combining conll construction data entity entropy filtering framework from hwee kubota labeled learning leong lexicon lluis marquez maximum mitchell multiple named pages perceptrons phrase predictive proceedings ranking ranlp recognition references report semantic spectral structures tasks technical tong training unlabeled with xavier zhang
http://acl.ldc.upenn.edu/P/P05/P05-3.pdf	107	ACL-05	aaron acero aist alex alexander algorithm algorithms allen alon amanda anagha analysis andras andrea anna anubha approaches ariel author automatic banerjee based behrang belvin beth brian buntrock campana carlo carol categorization cathy catizone chambers chan chang chatzichrisafis chelba chen cheng chia christina christoph chung church ciprian crescenzi csomai cues daniel data david dempster deneefe devault discovery documents duffy elements elkiss ellen emil engineering ettelaie evaluation extraction farrell ferguson foster from galescu gandhe george georgiou gertz giansalvatore goodman gregory guizhen gurevych hayward hearst hendrik hideharu hiroshi hitoshi hockey hongfang hongjiang howard html hyeon icdar icde iepad incomplete inderjeet index information inrt iris iryna isahara james jang jason jayaraman jean jessica jian john joshua journal jung kariaeva kenneth kevin knight knippen koller kothari kulkarni kuniko laird large lavie likelihood litkowski littman lucian manabu mani manny marc marcu marti mary masaaki masao matsuo matthew maximum mecca merialdo metrics michael michel midori mihalcea millward minoru mohit mueller mukherjee myung nagata nakagawa nakajima nakov nanno narayanan natalia nathan neel neely nichols niederlich nikos okumura oliviero oved page pages pakhomov panayiotis paolo parsing patrick pattern patwardhan pedersen philip phillips preslav proceedings pustejovsky rada ramakrishnan rapp rayner rebecca references reinhard renders repetition resnik reverse roadrunner robert roberta roser royal rubin rumshisky saikat saito satanjeev sauri schwartz scott semantic seok serguei series setzer shao shrikanth shyamsundar siddharth sites society statistical stefan stent steve stock stone stoness strapparava structures structuring sudeep suguru sundaresan swift tanimura text thater thiesson tomoyuki towards tracy traum utiyama valter verhagen visual vldb wenfang white wolf yang yiming yoshida yoshihiro yudong zhang zhangzhi
http://acl.ldc.upenn.edu/P/P05/P05-3003.pdf	111	Efficient solving and exploration of scope ambiguities	actual alexander algorithm algorithms allow already also althaus amsterdam annual appeal association available avenues been between bodirsky bridging broadcoverage budapest butt carl chart christian cken coling collects colloquium computational computes conclusion conference constraint constraints context contradict convenient converting copestake could count counting criterion data debugging denys description descriptions developers development different discrete dominance duchier dynamic dyvik eacl efficient eliminate engeneering english enumerate enumeration environment ernst evaluation explicit explore figure first flickinger formal formalisms formedness from functionality future general grammar grammars graph graphs have having helge hiroshi hole holloway however hpsg http inferences information interface into introduction ivan joachim johan journal kallmeyer king knowledge koller kurt lambda language large laura level lingo linguistics lisbon logic ltag manuel maribel markus masuichi means meeting mehlhorn miele minimal miriam more most nets niehren normal obvious open opensource operations output packed pages papers parallel perspective place pollard possible practice predicate presented press proceedings processing produced project provides readings real realistically recursion references related relatively representation representations research resources results rohrer romero rondane runtimes saarbr scale scope sebastian semantic semantics several siam small solution solver stanford stefan still structure structures such supports sven symposium syntax system take tenth testing that thater theoretical these thiel this thus today tool tracey treebank underspecification underspecified unification unplugged used useful user using utool vancouver variety weakly well whether with without workshop world years
http://acl.ldc.upenn.edu/P/P05/P05-3011.pdf	119	SPEECH OGLE: Indexing Uncertainty for Spoken Document Search	acero alex analysis anatomy applications approaches arbor audio been boston brin chao chelba church ciprian computer conclusions data developed engine eurospeech future geneva glass going have hazen hetherington hidden hltnaacl hypertextual ieee indexing interdisciplinary investigations isdn james june kenneth language large lattice lattices lawrence lecture markov massachusetts michigan models networks page pages position posterior preliminary proceedings processing rabiner recognition references representation retrieval scale search selected sergey specific speech switzerland systems timothy tutorial volume wang ward where work workshop
http://acl.ldc.upenn.edu/P/P05/P05-1004.pdf	5	Supersense Tagging of Unknown Nouns using Semantic Similarity	academic acapulco acquisition adwait alberta algorithms american amsterdam andrew anita annotated annual answering applications applied approach approaches artificial association australia automatic baker based beatrice beeferman better bodenreider boosting boston boundaries brants budapest building burgun cambridge canada caraballo carroll categories chapter charniak christiane chunking ciaramita clark class classes classification clustering college collocation columbus combining comparing computational concepts conference content context corpora corpus crammer curran customizations customizing darren database david determining developing development disambiguation discovery distributional domain dominic doug douglas editor edmonton electronic empirical engineering english enriched entropy estimation eugene european evaluate explorations extensions extraction fall fellbaum fifth france from grefenstette gregory grok guido harabagiu hearst hierarchical hierarchy hinrich hofmann human hungary identifying improvements information informative intelligence intelligent international investigating james japan jeffrey john johnson joint june kluwer knowledge koby koeling language large learning lexical lexicon linguistics lisbon logic marc marcinkiewicz marcus marius mark marti mary massimiliano maximum mccallum medical meeting melbourne methods mexico minnen mitchell models moens montreal morphological morton multiclass nantes natural netherlands network north nouns number olivier online open opennlp other pages park part pasca pearce penn philadelphia philip pittsburgh portugal press probabilistic probability problems proceedings processing publishers quebec question ratnaparkhi references research resnik resources retrieval reynar roget role sanda santorini sapporo scaling schutze seattle semantic sense sentence series sharon sigdat sigir similarity singer smoothing space specificity speech stanford statistical stephen suit supersense symposium synonymy syntactic system systems tagger taggers tagging task taxonomies technology terms text theory thesaurus thomas thorsten tokenizer toolkit trained treebank ultraconservative unified university unknown unsupervised usage using very washington weir widdows with word wordnet words workshop world yarowsky yoram
http://acl.ldc.upenn.edu/P/P05/P05-3022.pdf	130	Two diverse systems built using generic components for spoken dialogue (Recent Progress on TRIPS)	abstract advisor agents aist allen annual assistant august autonomous blaylock bologna byron calo carts chambers chicago choosing cocosda cognitive collaborative committee computer conference coverage data databases dept dialogue donna dzikovska expanding familiar ferguson filling first from galescu gaze george http human international investigators island italy james jeju joint july korea learns linguistic lucian mark mary meanings medication meeting member mining model mouse multiagent myrsolava nate october ordination organizes painting perrault placing preliminary principal problem project references report robert rochester rotating science sentences shen society solving speech spoken standardisation swier swift system systems technical that university vending virtual website with workshop xipeng