http://acl.ldc.upenn.edu/C/C00/
COLING 2000
http://acl.ldc.upenn.edu/C/C00/C00-1011.pdf	11	Parsing with the Shortest Derivation	aaai about aclcmans adjoining advam aiilslerdali ajman algorilhin algorithm alld alsguage altguag alwing ambiguity ameworl amsterdam analysis angtta angttctge anguagc anguage annotaled annotated approach araballo architecture arpa arsing arti arttl ased automatic bank bcnjalll berg berlin best beyond biased bilexical biology black bonncma briscoe buying cambridge canadzl carlo carning carroll chappclicr charniak chart chcqjter chitrao chrislodoulakis cial clmologies cnmark coijnf coling collins colloqttittm colloquium comlmtational comlmtatioual comlnaatioual communication comparing complcxity contcxl context copenhagen coping corpora corpus corpusbased corpusl coverage csli cubic current cxical darpa dings dissertation dopi eaci earni earning eclure eisner empirical english epoit erience estimation estimatkm european evaluation exicalizcd extending extraction ficalion figures first flickinger flonl fourth france free frolll ftmctional generalive generalized germany ghtcd goodman grammar grammars granlnlars granunars grishman grove gttistics hard hlspircd hmrnal icient iermany iipsg ilalnaparkhi ilased ilistic imii imxicalizcd inconsistent ingttistics inlelligence inlerpretation insertion intelligence international interpretation iota isambigua isambiguation issue johi johnson journal jtrcdlt kattfinann laillloll language lanllnars large larwu lexicaliscd lhtikling linguistic linguistics linlropy linq lintropy lish llead lllllll lllllllals llonncma lnlelligence lnside lnsme loogweg loughton lsstws ltlel lume lybrid machit madrid magerman march marcinkiewicz marcus maslel mass mathematical maximization maximtm maximum mceedings means memmy memorybased menlo mercer merit messages method mguage mifflin milkov minimality model models monte montreal morgan naaci nantes natmcd natttral natural nclherlands nculnann nethcrhmds netherhmds netherlands neumann nicolov ninth obus oceedings operation optimized oriented oslon pacific pahnucci palwing park parse parser parsing pcnt pearl pennsylvania philadclplaia plobabilistic press probabilislic probabilistic probal procedure proceedings processing publicalion publications quantitatively questions ramework rammcuw ramshaw recent redundancy references related ricntcd rivcn riven robabilistic robalfilily roblems rocc roce roceedings roceedinw roceedittgs rocessing saarbriickcn saartwiicken sampling santorini scha schwarz scnttlc semantic sima slochastic solits spain special speech squib ssing stalislical statistical statistics stochastic submitted sycho syntactic system technical theoretical theory theot thesis three through time trcebank tree treebanks uistics university unknown using utrecht utsidc very vperimental washington wcischedel wilh with word words workshol workshop xlwrimental young zipf
http://acl.ldc.upenn.edu/C/C00/C00-1059.pdf	59	Corpus-dependent Association Thesauri for Information Retrieval	accurate acknowledgements acquisition agency agglomerative allows also analysis analyzing application applications applied approach article artificial association associations automatic automatically based between both briefcolnparison browsing cational center church cluster clustering coincidence coling collection collections colloo compares comparison component compound computational computer conclusion cone conf consists cont context cooccurrence corpora corpus correlation croft cutting data demonstrated developed development digital disambiguating discovery doctnnent doculnents document documents dunning efficiently elicit enable evaluated even experiment experiments exploration explore extracting feasibility finer from functions gather generated generation grefenstette hamdouchi handling having hearst hierarchical hnage honkela hypothesis includes industry information inlbrmation intelligent interactive international japan japanese jing jipdec journal karger kaski knowledge kobayasi kohonen lagus large lexicography library linage linguistically linguistics lists mainichi make maps marchionini method methods mining ministry miti moreover multi mutual navigator needs networks neural newspaper newspapers next norms nouns novel ofhyponyms organization organizing overview part pedersen permitting pported previotis previous proc processing produce project promotion proposed proved putational quite reexamining references related research response results retrieval retriewd riao roms ruge saurus scatter schutze self selforganizing semantic shimbun sigir soergel state statistically statistics structure such surprise syntactic system tanaka tandling technology term terms text than thank that therefore thesaurus these they this through throughput time tokunaga topic trade treat ttanks tukey using vague very well when while willett with word work zooming
http://acl.ldc.upenn.edu/C/C00/C00-2131.pdf	132	Finding Structural Correspondences from Bilingual Parsed Corpus for Corpus-b sed Translation	alignment analogy analy anslation anstation approach artificial august banerji based between bilingual chine coling computational conf conjunctive context corpora corpus correspondences detection driven elithorn english example examples extended extracting figure forests framework free from grammars grishman human iida info intelligence ishimoto issues iterative iyanslation japan japanese june kaji kids kohyama kurohashi large learning linguistics linguisties long lyanslation machine manipulating matching matsumoto mechanical mecord memory method methodological meyers morimoto nagao nato parallel pattern patterns presenting principle proc processing proe references sato sentences shared similarity slot structural structures sumita syntactic system takeda templates texts theoretical tion tool toward transfer translating translation utsuro vatanabe very watanabe with workshop yanharber yanslao
http://acl.ldc.upenn.edu/C/C00/C00-1024.pdf	24	A Multilingual News Summarizer	according adev adopted advances affect agments also ambiguily ambiguity amta analysis approach arti asia association attd auroral automated based bian bloedorn broil bulgaria california canada chains chark chen chert chinese classification coling company computational computing concept conference conqmtational contextual corn corpus cross cues date decay defines degree denote descriplion developed different discuss documenl effective elements english erence erenee expressions ferenc first five fllture formats fourteenth frequent from functional future generating gives giwmmar graph helps high hmet hmguage hovy however http huang hybrid icial identification identify iian implicit importance index information inli intelligence international into kong language languages less lhmetuation line linguistic linguistics link linking lisrence litelwl lqformation ltong mandarin mani marcu mareu marks matching mckeown meet meeting message miss misses models monetary montreal multi multilingual multiple name national natural news nglish nguages nouns only oxford pair pairs paper partition performance pivceedings pjweeedings polysemy predefine preferred presentation presented press proceedings processing proper proposed provided providence ptvceedings publishing punctuation quebec quit reading recent reduce redundancy references reported reporters requirements research resolving results retrieval retriewd rmation saic salne search select sentences sequence shorter should shown similar site sites smmnahzation some sources stories strategy studied summaries summarizalion summarization sunnnarization system tanding target terlns text texts things this thompson tian time together topic transhtte translation transliteration tried tutorial tzigov undel unique university used user voting which with workshop yang zone
http://acl.ldc.upenn.edu/C/C00/C00-1001.pdf	1	A word-grammar based morl)hoh)gieal analyzer for agglutinative languages	academic actd aduriz agghttinative agirre alcgria aldczabal alegria analysis anltila ansa ansdttcers application approac approaches arabic aradic arlola arrcgi arriohl arriola arsitl artola artolax athens auiguagc automatic based basil basqtte basque bcesl blackwcll cambridge cicyq ciencia clmrtmcnt coiang comlmtational comptttational computcaion computing concatenative confcncncc conference constrai context cotfforence cotllisidn cotlntry csli design dfaz documehls ecciza english epetme eral estricled ettcoding evaltmfion evaluation ewduation exico ezciza ferelt ffng first fjlack flazcr form framework from general german gnmmmr gojcnola gqicnola grammar grammmar granada greece gruyicr gtlagc hascd hclsinki heikkiht hhtiatire highly hudson iangual ianguisiics ianguistic iatcrary icnohi ilarraza ilclsinki inite inlernational insau integration international irst issues karlsson kartmnen kgmtmd kluwcr koskcnniemi lall language languages larraza ldcpcndent lemmatizer level lexica lglish lhlglish lnjlectional lnlernet lntcrministorial lntroductiotz logy marilxahu maritxahu maritxalar maryland mechanism model moqdto moqdzological moqgtology morldzo morphological morphology moudu mouton mttlli mulii mullilingual natural oronoz other oxford parsing pmce pmduclioth practical press proc proceedings processing proposal publishers pulhnan qdlo qtte raza recognition references resources ritchie roach rocccdings rtended russcl rzon sarasoht sarasola second series sgmllagged shicbcr sigi slanford soma soroa sproat state sttl syntactic system tagger tags tcxls tecnologfil ternational text tmgei tools treatment trost tructin ttrposed turf ublications ublin unificati unification university urizar urkia using vcronis vical voulilaincn with word workshop yhmd
http://acl.ldc.upenn.edu/C/C00/C00-2172.pdf	172	Construction of a Hierarchical ]¥anslation Memory	aine aligmnent anslation applying asexl august based bilingual brown casca cascaded category college coml computational comtflel consisting contract convert copcntm corpora cortms coverag data dialogs ditficulty edings emnark empirical especially exmut fede fimflly finland first focus futur german good grammm handcrafted hclsinki heme hnoh icial inchuh ings integrated international into itmjo joint june land language large lattice lingui linguistics lion litt lmprove lmrformcd lnlmls luce machilw machine main mcccdi memory memorybased mentioned meth mguag ministry mlse mobil modal model models morgmt mory mrea nagao natio nfere npcr nsducers nulnl onsistenl orted othes ould pangloss park partly phis presen procceding proceedings pronch ptwccs rans ransb ransla ranslation rcncc rcnce recognition references rehensive rencc rent rmlsdut rmmdu roach roadl rocc rodu sato scoring seal shown simple spee ssary state statistical summit systeans system target tchine tclligc tcrn test tested then this tics tillmmm tion tirs tngmtge tnmslation tnrget towmd transdu translation trmlsla trmlslni ttion tween ufmmm under university used vert very vohmm vonu wahlster which will word work wral
http://acl.ldc.upenn.edu/C/C00/C00-2110.pdf	111	Japanese D e p e n d e n c y Analysis using a Deterministic Finite State Transducer	acknowledgment algorithms alld analysis analyzer anlp appreciate base brisawa case coling conducted construct corpus crafl cxicalizcd dctmndency decision dependency domain effort emmmmel emnlp eriments experiment finite fujio grmnmar hand haruno helped hybrid ichi japan japanese kanayama kcntaro kindness kurohashi kuroluu kyoto lalmnese language like liroshi making makoto masahiko masakazu matsumoto means mitsuishi nagao natural oceedings oche ooymna other paper parser parsing people practical prepare procccding proceedings project publit pwcccdin rcnceon really references rocccding sadao satoshi sharable shirai shop sourccs state statistics strtlctlll structure text thank their this transducers tsujii uchimoto university using were with worked would yoshiflmfi yutaka yuuji
http://acl.ldc.upenn.edu/C/C00/C00-1051.pdf	51	C o m m i t t e e - b a s e d Decision Making in Probabilistic Partial Parsing	aaai aagc abilities academic algorithm analysis analyzer anese anlp annual aphasic apptvach arccs arsing artificial assist assistive association average based bayesian better bigrmn brill building carroll case charniak classifier classifiers cmming coling collins combination combining confcrcncc consistency contextfree corpus daelemans data dependencies dependency devlin disambiguation dismnbiguation diversity driven eacl ederking editors ehara emnlp empirical english ensembles entopy error estimating estinlating eurospccch evaluation exploiting extreme fiscus fujio generative grainmar grammar halteren haproved heads heidorn henderson hpsg httcrnational integrating intelligence isahara japanese japmmse jensen joint kanayama kluwer kurohashi lang language lexical lexicalised lexicalized lical lint matsmnoto maximam maximum meeting minnen mitsuisi modeling models naacl nagao naive natural networks newspaper nirenburg nlprs ostcrior output parsed parser parsers parsing pedersen plnlp poole post practical prior probabilities proc processing prvc publishers rates rcso readers recognizer reduced reduction references relations richardson roach rover roving search sekine sense sentences sharablc shirai sigdat simple simplification statistical statistics structure system tagging tait tanaka technology text three tile titan tokunaga torisawa trot tsujii uchimoto using voting while with word wordclass worksh workshop yield zavrel
http://acl.ldc.upenn.edu/C/C00/C00-2171.pdf	171	Incorporating M e t a p h o n e m e s in a Multilingual Lexicon Carole Tiberius and Lynne Cahill	above acquilex adjectives allophone architecture association automatique auwera baayen books britain cahill celex choice coded community computer concept conclusion conditioned connor consortium contlmtational copestake correspondences cross database datr determiners discussed english envi esprit european evans final gazdar gemini gerlnan germanic great have individual inentioned inllectional innovation international jones journal kfnig knowledge language languages level lexical lexicon lexicons linguistic linguistics ljtnguages lnultilingual london lower marinai metaphoneme metaphonemes mgues mlex monolingual montemagni multifunctional multilex multilexconsortium multilingual notation paper paris particular pelican pennsylvania philadelphia phoneme phonemic phonetic phonetics phonological phonology piepenbrock polykex pronouns realisation references related release report representation rijn rodriguez routledge sanfilippo standards such this university vossen wells witement working
http://acl.ldc.upenn.edu/C/C00/C00-2147.pdf	147	The W e e k at a Glance Cross-language Cross-document I n f o r m a t i o n Extraction and Translation	although amount appears applications approach area artificial association based basic calilbrnia canada cling computational conclusion conference coverage cowie current demonstrates development diderot documents domain eltort entity erence extraction feasibility filwt flom fourth generate guthrie harman hnproving independent industrial information infornaation initial intelligence international issues italy joint kaufinan kaufman knowledge laboratory language linguistics ludovik mahesh mateo meaningful message mmnce molina moncton mono month montreal morgan multilanguage multilingual named natural nirenburg nist ontological ontology pacific particularly perlk possible practical proceedings processing publication pustejovsky reasonable recognition references required resource retrieval robust salgado sharing significant situated sixth sources special summarization system systems tanding text that tipster trec under vancouver venice vextal wakao waterman workshop zacharski
http://acl.ldc.upenn.edu/C/C00/C00-2151.pdf	151	An Experiment On Incremental Analysis Using Robust Parsing Techniques	academic account acknowledgements aitemcnt alain allow also amework analyse analysing analysis andreas annual anytime appears applications applying approximation artificial automatique based because been bennett berlin canada case caused certain change chartparsing cheap clear coling comes communication computational computer confer conference conflict conflicts consider constraints contribution coupled crucial deciding decision degree delaware department dependency dernom description deterministic deutsche dialogs dienom editors eech eine eliminative ence endance enge england enough erlangen estimated european extended face fast fimded find finland first forschungsgemeinschaft foth foundation frau german germany given good graded grammaires grammar grammars grant hans harold hauenstein have heinecke helsinki here heuristics idea incremental individual information ingo inkrementelles integration intelligence interfaces interna international interpretation into investigation irgen italy itans iwpt johannes johnson joint journal kahane kilian kopplung kunze language langues later licensed linguistic linguistics linksping lombardo london maintenance mama marcus mary mats mcgee meeting menzel michael might mitchell modeling montrial montrsal more national natural naturallanguage necessary newark next niirnberg obtained only optimal oriellted ossible pages paper parser parsing partly paul peter play polgu precisely preference press probabilistisches procecdings procedures proceedings processing property proximation reason references refined required research resembles resolution restricted results right roblenl role schrsder schulz science search seattle sees seiners should sieht since space speech strategy structural structure studies subject subsentence subsequent substantially such suche sweden sylvain synchronous take technique techniques technologies technology ternational then theory there therefore this tightly tile timal time tional tlmsis transformation transformations translation trento under unification unifikationsgrammatiken universitat university using vceedings verbmobih verbmobilreport very vincenzo wahlster washington weak weber when which whitclock will wirsn with within wolfgang woman wood word workshop worthypothesengraphen wrongly
http://acl.ldc.upenn.edu/C/C00/C00-2145.pdf	145	A M o d e l of C o m p e t e n c e for Corpus-Based Machine Translation	adaptation aides aiming altay anslation approach apptvach asscociation austere automated based bcnutzcrhandbuch bochunl boitet brown call carl cbmt characterized christian cicekli coarscmuiucd cockc coling college collins complex computational computed computer conclusion connectionist della democratize dichotomies dictionary dublin dummett enable european examined example examples extraction final fine form formalized free from ftbw geneva germany giivenir gmbh grained graphcnfic guided halil hcyn high holistic holistie ilyas implementation including inducing information integrating into issco jelinek knowledge language learning linguistics machine makoto mapping matching matthias mclean mean meaning memories memory memorybased mercer mere michael mind molecular morphological nagao nlachine nmation number order output oxford pages perform perspective peter pietra preserving press proceedings provided quality raapping ralf references relics representations research respect retrieval rich robert roossiu sato sernantical source statistical stephen such summit surface syntactical system systems systenls target templates text that theories theory these thesis they three towards translation trinity university uses using vincent what while with workshop zeres
http://acl.ldc.upenn.edu/C/C00/C00-2105.pdf	106	Robust German Noun Chunking With a Probabilistic Context-Free Grammar	abney academic according additionally algorit algorithm algorithms alize almey anne annual applied arison arses artificial asking atro augmenl automatically automaton bascd base based berwick bilexical brants carol carroll cascaded cascades case category ccedings charniak chelba chris chunk chunker chunking chunks church ciprian clmnker collins coml computation computer concordances confcrence conference context corpora cover data dependencies dependeney design detector develoi different dmor ected editors efficient eisner elled empirical english essen estimation eugene evaluation even exploiting extensions feature figure fiir finitest franz frederick free generative german giorgio glenn goodman grammar grammars granunar grmmnars head headlexicalised hehnut herlnann hess hnplementation hunker identil inarkov induction ined ings initial input inside institut intelligence international jason jelinek joshna joshua journal jsrvinen kenneth kluwer kneser label language large lari latye learning lexi lexicalised lexicalized lexicography like lishers lopar manuscript marcus maschinelle mats maximises mceti meeting methods metrics michael modeling modelling models national natural noun nptool only ooth order outside pages parsing parsitt partial parts pasi pcfg phrase phrases prec precision pree prescher probabilil probabilistic probability procccditt proccssing procecdings proceed proceedings proceedit processing progranl ract ramshaw range rather recall recision references reflex report resenl resulted riezler robust rooth rules satta schiller schmid second semi shop single specch speech sprachverarbeitung spraehverarbeitung sslli statistical statistics stefim steven stochastic strate strategy structure structuring stsckert stuttgart syntactic tagging tapanainen tciplc technical technologies tenny text third thorsten three time trailfing training transtbrlnation universit unlal unlcx unlex unlexl unpublished unrestricted untrained using valence vcry voutilainen were which with word workshop young
http://acl.ldc.upenn.edu/C/C00/C00-2114.pdf	115	A Dynamic Language Model Based on Individual Word Domains	accurate acoustic alpha also amsterdam analysis approach april automatic average bahl based basic best biometrika boyle cache called capture colorado combinations combined combining component computer concept conclusion continuous corruption data decay december describe different domains entropy estimation even explored exponential first frequencies from fully gelsema global good gram hanna have hence hnprovement icassp ieee improve improvement indicate individual intelligence interpolated introduced jelinek kanal katz keystone kuhn language likelihood machine markov maximum means mercer method ming minneapolis model models mori natural norlhholland obtained occur offer order over owens pairs paper parameters partial pattern performance permit perplexity population practice preliminary probabilistic probabilistie probabilities processing products promising publishing recogniser recognition reducing references respect result results rosenfeld roukos show signal significant smith source sparse species speech stewart sums table temporal that they this though transactions trigger understanding union value weighted when which with word words workshop
http://acl.ldc.upenn.edu/C/C00/C00-1074.pdf	74	Hybrid Neuro and Rule-Based Part of Speech Taggers	acquire acquired aelemans aykin based been berck brma building cannot case cilll coling colingacl college collstruct computational computer constructed contexts copenhagen coral corpus correcter crucial curacy data dealing driven eech english error even examining experiments found generalizing generating ging good hardly have high hose however hybrid ijcnn ility including information inputs introduced isahara japan kaufinann kyoto language large lastic learn learning lengths lexical linguistics little lsahm machine macmillan make mateo meriaklo mlory model montreal morgan multi mura natural nchvorlcs needs networks neural neure neuro nich nnot obtained ofda pacific parl part phuket possible post prec probabilistic problems processi processing programs proposed publishing quinla recessing references rill rning roenporll rseness rule rules sahara schmid shortcomings single sornlertlanlva speech study system tactical tagged tagger taggergenera taggers tagging template text thai thailand that them therefore these they those tion training trans transfornmtion transtbrma trouble uchimoto uistics using utilize variable very washington were when which with workshop wrel
http://acl.ldc.upenn.edu/C/C00/C00-1044.pdf	44	Effects of Adjective Orientation and Gradability on Sentence Subjectivity	aaai adrienne albany alld american amster analysis anlp anslation applicatiolls applied artificial association augt augusl aunotaled austin base basic bates battistella beatrice bruce building california cambridge canada case ceedings chiet church classilieations cmguage cobu collins coml complvhe computational conference corpus culture data david delbat development diane dictionary discrete edited edition editor edwa edwin eech elds england english etwtructure evahiative evahtalien february ference fleiss fmglish geoffrey gineering gold grading grammar granada grecnbaum guistics hara harper hilosol holland intelligence issues jall jerrold john joseph july june kalz keley kenneth kevin knight knowledge language lanyce large largescale lehrer lexical licai linguistics lntenmtio london longman lrec lume machine mahesh marcinkiewicz markedness mary maryhmd meeting methods milchell montreal mportions natio natural nirenburg nomi north noun ogranl onald ontological ontology ouglas pages paris parser penn personality phrase pmceedi practical press proceedi proceedings processing qfscie quirk rates rebecca recognizing references regression reprinted resources rllanual sanlner santorini sapir sealtlc second selected sema semantics sharing sidney siluated sinclair sive spain springer standard state statistical steve stochastic structztre study subjectivity svartvik tagging tandolph telli texas text theory thomas tics tile tional total treebank ttaward tttatioltal uffy uistics universily university unrestricted verlag vlarcus volume washinglon watts wiebe wiley wiqtings workshol yons york
http://acl.ldc.upenn.edu/C/C00/C00-1068.pdf	68	Flexible Mixed-Initiative Dialogue Management using Concept-Level Confidence Measures of Speech Recognizer Output	accel acceptance aiifl annual arise arser ased attribut audio automatic balancing based bouwman boves busayapongchai category colnnmnication colntmter comlmter concel concept conf confirmation confirmations confirmed confl constraint content contidcnce contidence continuous contirmation control criterion decide detection developed devillers diah dialog dialoguc dialogue domain doshita doshitm dutch eech effective eilbx error errors esca estimating euil evaluating experiences false flexible form formation generate glass grammar guage guidance guidances hmid hockey hone hrase icassp ieee ieice illteractioll incort independent ined information initiative intention interactive interfaces interpretation isard issues juang kawahara kearns knill kobayashi lamel language lapancsc lapau large latform lean level line litman maka make mallagement manager measure measures meeting meng mixed mlderstnnding mnacef modal moreover multi niimi onfidence onses optimized orating order ospeeci ossen pages pair plications pokc pokcn poken polifroni predict present proc processing propose provides query questions realize realized realizes recognition reduced references rejected rejection reliability resl rocessing roject rolmst rosset scoring search semantic seneff serr setting should simulation society speech spejewski spoken stone strategies strategy stuff sturm syst system systems telet that this thresholds timetable tive tlexible train trans ttnns tvcc understanding user using vcessing verification vocabulary walker watanal whether which with word workshop
http://acl.ldc.upenn.edu/C/C00/C00-1065.pdf	65	Hypertags	about academic ahnost almost ambiguity amsterdam application approach benjamins clin coling college company complexity computational derivation description descriptions disambiguation domain eacl exploiting explorations froln generalized generation grammars ibingen intelligence john joshi kalhneyer kinyon kluwer kyoto language lexical linguistics locality ltags manaster masaru model obtaining paris park parser parsing partial parts pennsylvania philadelphia prec preferences proceeding psycholinguistic publishers publishing rainbow ramer reductionistic relevance remarks representations resolution rogers shanker some speech srinivas super supertagging supertags tags tapanainen their thesis tomita trce tree trees undcrspecified univ universit using utrecht vijay voutilainen with
http://acl.ldc.upenn.edu/C/C00/C00-2108.pdf	109	Chlstering Verbs Semantically According to their Alternation Behaviour	according acqui acquiring acquisition aims algorithm alternation alternations analysis anand ancesc annals annual appropriate apptvach associates association autoinatic automatic automatically bangla based beckwith behaviour berwick beth bnild bonnie brian building butt canada carroll case ceedings chaptcr chapter chicago christiane class classes classification clusters comp compositional computational computer confcrcncc conference copenhagen corpora corpus csli cues data database dempster derek diathesis dimensional disambiguation dismnbiguation distributions doeulnent dorr doug douglas dublin edition editor editors eitung empirical english erlbaron european eurot experiment exploiting factors features fellbaum ffir francesc franklin from generalizations george german geuder glenn grammatical granada gross head hflginfl hillsdale hovav hughes incomplete inducing induction infl information institut institute intcrnational international ireland jersey john jones journal judith karen khan kohl korean kullback laird language lapata lasses lawrence lcxical lczicon learning leeds leibler levin lexical lexicalized lexicons likelihood line linguistics llow malka maria maschinelle massachusetts master mathematical mats maximum mccting meanings memo merlo methods miller montreal naoyuki natu nomura organized pages paola parsed pennsylvania philip predicting preference press principles procecdings proceedings processing psycholinguistic publications pwcccdings pwceedings radhakrislman raehverarl rcncc references relations relationsh report resnik resources restrictions ribas richard rmation robert role rooth royal rubin sabine sauerlan school schulte selcctional selection selectional selectionm semantics senmntic sense siglex sition society spain sprachverarbeitung stanford statistical statistics stevenson studies study stuttgart sufficiency suzamm syntactic technical technology their thesis ulicny ulri universitgt universitst university using utational valence vans vcccdings verb verbs walde with word wordnet words workshop zeeshan zernik
http://acl.ldc.upenn.edu/C/C00/C00-2143.pdf	143	Dependency Treebank for Russian: Concept, Tools, Types of Infornmtion	acquisition aductique adverbial amlolations analytical anguagc anguage another anslation apples aprcsian apresjan atttomatique auxiliary azurskij been between boguslavskij bouillon brings building buiming cases clas clauses closed come complctcd computational computazionale conclusion conference conj conjunctions constructions coordinalive coordinated coordinative corpus covered crcation data eebank electronic elements elext encoding enden english etap evahmtion evaluation evening first five flfll form generated grammars gtirly gtte guage guidelbws guidelines hardly hashi html iiuman inlernational international introduction iomdin japanese language large lazurskij lfinguistics lhal liaiicova linguistica linguistics link list lmprovbzg machine make marcinkiewicz marcus markup meta modifiers monlrdal montrdal nagao need noun opinion order otated pages panevova parenthetic parsed parsin peajwlyl penn prcscnt predicate presses proceedings process quanlitalive quantitative quantity rare really references reflect relation relations relationshil relationships resources reusable righi sannikov santorini semantic semantics serve sgall single some state stbme structure such survey syntactic system technology tetwhange that them this tile tmduction traditional tsinlnan tsinman types typically unambiguous unctive unit universitd variety varile various vebank virginia while will with word words xiii zampolli
http://acl.ldc.upenn.edu/C/C00/C00-2099.pdf	100	A Statistical Theory of Dependency Syntax	ages aktcn alshawi also analysis anliifllich annual applied arbeiten association automata avid based basil bilingual black blackwell brel cambridge christer codes collins complexity computational confcrencc confercnce context control crnando csli david decision dependency deulscl disambigua ecognition engel england ependency ewan ezra formalism franzssischen fred gaiflnan gazdar gebursttagcs generalized generative geoffrey gerald german gorn grammar grammars harvard history igor iiead ilays infinite information international itelsinki itiyan ivan jacques jane jelinek john jsrvinen khalil klein klincksieck kolloquiums lafferty languagc language languages lcmguage lecture lexicalized libraire linguistics linguistische lishing llelsinki loldmcnts ltaim lucicn lucien magerman math mcntalcs means meeting mercer micha miflvcrstanden mille minimal models natural niedermcyer nonprojective note obert observations ojnli operation oxford pages paris parser parsing pasi pereira pgvcs phrase piii press probabilistic proceedings processing processors procs proes prolog published pullnm quitte rcworl references representations richer robinson roukos rules rvinen salim samuelsson saul seines shannon shieber sima some springer state statistical stochastic structuralc structurc structure structures stuart syntax syntaxc syrup systems tapanainen tcsni tcsnidre temps tesnihre theory three tiibingen tiling time timo tional tions towards transfornm translation trasbourg tree treegrammars type uctural ucturalc ulrich university using valse verlag very volume wiedcrkchr with ynta york younger
http://acl.ldc.upenn.edu/C/C00/C00-2140.pdf	140	DIASUMM: Flexible S u m m a r i z a t i o n of S p o n t a n e o u s Dialogues in Unrestricted D o m a i n s	aaai access accessing aceessin achel acoustic acts advances aibel aist alex alexaudersson allen alph ambridge american ance andreas andrew anih anlp annotation april aren arima arizona association atltonlatic audio aurafsky australia automatic auzanne available awfilablc barzilay based bates belt bert best boguraev braniinir brill broadcast browser buckley budapest building camloridge canada carbonell carol cedric ceedings cgina chains chapter characterisation chris christine chrlstol chuni chunk classification cnnedy coccaro coling comn computational computer conference content conversational cornell corpus correction ctrieval daniel darpa decen department detection development dialog dialogues dilek disfluencies diversity docillllelll doculnent documents docunlents domains dykema eacl editor edufzechner eeci eeedin elhadad elizabeth ellen eric error esca evaluation extraction fformatio finke first florence gaithe garofolo gawddh generation gerard germ godfrey goldstein gregory gross gskhan guistics html http human ickcy icslp ieyung ihao iiakkani iiearst iieeman iiill iiongyan ilcl ilek iligh ilirsehberg ilugust indicators inetrics information inforniatiou infornmtion ings inlg intelligent international into introduction investigation isao islizabeth iuassp jade jaime james jing johl john jskhan julia kameyama kawai klaus lahc level lexical lexlble linguistic linguistics lists lizabeth llua lmiguage lnjormation loken ltetrieval ltolliman madelei march marianne marie marl martin martl master matching mcgill meeting meetings megumi melbour meteer mfiel michael minimizing modern morsel multi multilingual multimedia multimodal naacl nakatmfi narlzation news nhcatio niagara noah north november obin obinson oger oken ones ontancoijs oombining ooundm ostendorf overview pages paragraph paris parsing part partial passages paul pcchnical perforlnauce peter plauche plication poller procced proceed proceedin proceedings proceeedings producing progress project prosody protocol pubiications rate real rebecca recognition recognized recognizer reducing reduction references reordering repairs report representations reranking research results retrieval retriewfl rkshop rsburg salience salton scalable scheme science seattle segmentation segmenting segnlentation sentence sentences september shop shriberg sigijg some sparck speech spoken sponta spontaneous spontmmous spring stan stanford stolcke stratford subtopic suinnlaries summaries summarization summarizillg summarizing sunnnarisation switcttboard sydney symposium systcni system tagging telephone text texttiling textual tfir thesis through tinie tion tolcke topic topics towards track tracker tracking transcril transforlnation trigger tsca tucker university unrestricted using valenza vincent vohnne volunm voorhees waibel washington wiontreal word words work workshop xiaojin yang ylor zechner zeehner
http://acl.ldc.upenn.edu/C/C00/C00-1054.pdf	54	Finite-state Multimodal Parsing and Understanding	abella abney academic accepters aces addison also alure andrfis antiquity approximation aravind architecture august australia automata bangalore based berlin berwick bocchieri bolt bruce buntschuh cambridge canada carol carpenter chen chunks clew cltl cohen coling colingacl compilation complexio composition computation computational computer constraint corner coupling currently descriptions design devices difabbrizio directory disambiguation distributed ecember editol editors embedded emmanuel engilzeering england errors esigning exploring extend extended feature fernando finite fiom focs free fsts general generation gesture giuseppe gralnnlars grammar grammars graphics hiladelphia hmmut hopely httelfiwes human icsli icslp ieee ikovic information integration intelligent interaction interactive interface inultimodal investigating johnson johnston jollnson joshi kamm kaplan kaul kluwer kornai koskenniemi labs language large lecture left level lexic lhfite librao linguislics linguistics lizhong lnter logic machine madrid mann maps marcus mark massachusetts maybury mcgee mehryar michael model modeling models mohri montreal morgan morphology mullimodal multilnodal multimedia multimodal mutual nalttral narayanan natural neal nmltimodal notes number ovialt oviatt ozen pages palwing parser parsing partial pennsylwmia pereira performance philip phonological pieraccini pittman pmwing press principle proceedin proceedings processing processitlg production publishers rational readings recognition references regular relevance report reseamh riccardi riley robert roche ronald rosenberg rule scale schabes science sentences shaffer shapiro sharon sharp simulation sinki smidl smith spain speech spoken springm srinivas state statistical steven stochastic structures sulliwm sydney systems tape technical technology teclmiques tenny thal there thesis tight tlaesis transactions transducer transducers transforms translation tylm uncan unification unilicationbased university user using utational view voicc wahlster weighted wesley which wilpon with wordzform workshop wright york
http://acl.ldc.upenn.edu/C/C00/C00-2138.pdf	138	Identifying Temporal Expression and its Syntactic Role Using FST and Lexical Data from Corpus	abney about acadenlic accuracy accurately acquisition after algorithm alltonlata almey ambiguity analysis annual apcrs approach ascd assigns association ation ational automata average bank based berwick besides bigram boston bottom brants cand candidates cannot case caused cccdings chang chart chinese chnology choi chuifldng chunk chunking chunks ciravegna clmnking cntial collins compared computatiou conclusions concordance conducted conlbined consideration context controlling corpora corpus could crucial ctmnking data defined dependency description deterministic development develot different disaml discourse dismnbiguating does each earning editors effect efficient eftieient elgot endencies english enough ependency epresen errors esslli expression expressions extracted file finite flciency flmction from functioll functions future gale generalized generally global grmnmars gross guistics hand harpercollins have head histicated hoey hycordac ical iccpol identification identifies identify identit iguatil ilnprove impo infornlation ings instance intcrscnt international into itself john journal kaist klnwer korean kwon language large lavelli lexical llil local london losing luccrs makes manipulate manually marcus meeting ment method mezei michael model mohri most natural needed nevertheless njac nouns number numl nycon occurrences oche only otherwise paper parser parsers parsing part possible pracl present presented press principlc proceedings processing program project prol promising provide publishers purpose pwceed ramstmw rcscarc recognized reduced reduction references relations represent resolve resources ressions result resulted roche roles rsing same schabes sentence sentences sfive shih shown shows since sinclair sing skut sometimes song speech sproat state statistical stics stochastic such sults syntactic syntax system table tagger tagging tags take tant teinporal temporal tenlporal tenny test text that their them then this three through tian tile time tirst tomated tractable transducers transtbrmation tree treebank ttcrnational tycott types under used using very well were when where which whml whole with without wlfich word words wordsegmentation workshop year yeou ying yoon
http://acl.ldc.upenn.edu/C/C00/C00-1057.pdf	57	Robust Segmentation of Japanese Text into a Lattice for Parsing	acquiring algorithm analyis analysis analyzer applications approach backward best boston breaking broad brockett building coling conference connective consistent corpus cost coverage dolan first forward from fuchi heidorn hisamitsu hnprovements hnproving hypothetical ieice information input japanese jensen juman kacmarcik kitani kluwer kurohashi language lrec matsumoto method microsoft mindnet minimum mitamura modified morphological msrtr nagao nagata nakamura natural nitta occurrence only parsed parser parsing plnlp preprocessor proceedings processing records references report return richardson search segmentation segmenter segmenters semantic sentence signlc snlr stochastic structuring suzuki syntactic system tagged takagi technical text using vanderwende while with word zixin
http://acl.ldc.upenn.edu/C/C00/C00-2144.pdf	144	Thistle and Interarbora	academic annotating athens attd bernsen bkcnisation blumenthal calder cambridge chicago communication comparison computer conference context controlled convergence csli delivering diagrams dialogue direct discomwe discourse dordrecht driven editors erence evaluation ewhci finguistic flexible formalisms fotmdational from generating gornostaev grammatical granmtar greece grosse grover head hrase human hzteraction instructional interarbora joshi june kamp kluwer language lcmguage learning lncs logic london lsxues manipulation mate matheson maycs mekendree mikheev mildly minas moens moiler moscow natural nternct olsen pollard press problems proc proceedings processing references rence resources reyle roceedings russia schiffi science second sells sensitive shanker shieber springcr stanlbrd structure ternational thistle tobin tool unger university using verlag vicarious viehstaedt vijay wasow weir workbench ybkjler
http://acl.ldc.upenn.edu/C/C00/C00-1064.pdf	64	S t r u c t u r a l F e a t u r e S e l e c t i o n For E n g l i s h - K o r e a n S t a t i s t i c a l Machine Translation Seonho Kim, Juntae Yoon, Mansuk Song { pobi, j tyoon, mssong}@decemb er.yonsei,ac.kr	about according accuray acquisition adam adnominal ajma alex algorithm aliglulw aligmnent aligmnents aligning alignment alignments allj aman analysis anjay anseok approa approach approacll arpa ased ayaman berger bilingual books brown calldie categories chert choi chose church cocke coling collocations coml composed computational conclusion conditional conducted conference consideration corl corpora corresl cort dachero dangsin data decoding dekai della dempster determiner differences eacl embedded enco english englist ente entr entri entrol entropy equivalence eriment erty estimation example extend fanjayaimm fbatm feature features fields figure finding frank frederick fredrick frmlz from gain gale general gillett handa harry harvard hatziw have hiroyuki human ieee iivp ilans ilbansenk ilbanseol inatching incomplet information intelligence into ishimoto iyansactions jelinek jersey john josef julian jung kathleen kengo kenneth knowledge korean kupiec lafferty laird language large learning length less lexical lexicons lignments likelihood likely linguistics litiolml llel llnducing lubos machine mapping mappings marin martin mathenlatics matsmno maximum mckeown melame mercer method methods model modeling monotone more most natural nlachine nnde nndei nnin noun null oioi olynonlial ondenccs order para paralm parmneters partly paths pattern paul peter phrase pietra plainsborough pnyppca ppad ppau ppca press printz probabilities procccdinfls procccdings proceeding proceedings processing program ptvcccdings rameter randora reason recognition references regular request require result robert rocccdings roossin roscheisen roving royal rules sate seating seine send sent senten sentences sequence serving shin should shows silnlily smadja smarter society some speech ssiloglou stanley statistical stephen structural structures stuctural system table take takehito tatistics technology texts texttranslation than that thrown tiffs tilhnann time tlle translated translating translation tried ubin unit ures using usmtlly utational utsure vasileios vbma vincent vogel waibel wang were when will williain with word words yuji zubiaga
http://acl.ldc.upenn.edu/C/C00/C00-2152.pdf	152	An Integrated Architecture for Example-Based Machine Translation	acceptable algoritlun also analogy artificial associates baneiji based between brown cabulary carnegie cenler closer code coling components conlinuous constituent content csli dalryml dapest data developing direction driven editio efficient emcompiler engines english evalualc ewtlualion examples exanll extended extending fast feature finhmd fmrospeech framework france functional further furuse future generalized grammar guide helsinki holland hungary icassp iclionary iequirements iida include incremental index information input integration intelligence iranslation issues ivailure japan japanese jithorn kaplan lattice lcvine lecture lexical licai lratlslaliod machine major mason match matrix maxwell mechanisms mellon melnory memo memorandum memory memoryl minor nagao nantes north notes only ormal oundary pages parsing patlcrns pattern plan plans principle problem proceedings processing promem quality rcillcy read recognition references report resnik results roblcm roceedings sale sebastopol second semantic similarily similarity size slructures software speech stanford stylistic sugaya synthesis syslem system table takczawam takeda taxonomy technical this tlumatt tokyo tomita tonlila toward transfer translalion translation university user using utilizing version walanabe with word work working write wrong yacc yamamolo yokoo zacncn
http://acl.ldc.upenn.edu/C/C00/C00-1035.pdf	35	Aspects of Pattern-matching in Data-Oriented Parsing	algorithms alnsterdanl amlotatcd analysis annual antwerp antwerpen artificial association august based belgium bosch building cambridge canada combination competence computational computcrtocpassin conte corpus correct daclemans daelcmans daelemans data dissertation driven editors eebank efficient empirical english enriching fence fills fngland goodman grammar halteren improving intelligence istics jaarboek joshua journal kort landelijke language large lcerdam learner lingnisties linguistics lvvn manual marcinkiewicz marcus matehin meet memory methods model models montr montreal natural nccrlandistick ncerlandici pages papers parsers parsing pattcrn pauw performance perfornmnce perimcntal petal pmpg pond press probabilistischc proccssing proceedings processing quebec reference references report rpericnccbased santorini scha sloot stat system taaltectmologie taaltheorie tagging technical tgcvocligheid theoretical tilburg tillmrg timt univcrsiteit university vereniging version walter whnp with wordclass zavrcl zavrel
http://acl.ldc.upenn.edu/C/C00/C00-2135.pdf	135	Acquisition of Phrase-level Bilingual Correspondence using Dependency Structure	achieved acquisition aligning approximately bilingual brown business bussiness clues conducted corpora correspondence dependency dictio english experiment expressions find finding from fuller hashimoto have investigation japanese keizai language letter level linguistic mercer method methods mlingual nary natural nihon pages pairs paper parallel phrase precision preliminary propose references relations requires sentence sentences serve shimbun shows still structure takubo task that this though useful using vcessing with
http://acl.ldc.upenn.edu/C/C00/C00-2120.pdf	121	Matching a tone-based and tune-based approach to English intonation for concept-to-speech generation	above academic accents acoustics also among analysed analyses analysing analysis another approach approaches assignment australian aware ayers based basic basis bateman batliner beckman been behind benzlniiller between bitonal black bnes brazil british caley cambridge cassidy clear closely cmnbridge cole collaborate combine communicative communiciation comparison concentrated concept conditions conference consider consistency context controversy correspendences coulthard course cstr data descril description descrit deseription dialogue discourse documentation dordrecht edge edinburgh english enhanced equivalents equiwdents establish established existing experiences feet festival first foot from functional generation generator german goal grammar grice grote group groups gtom guidelines hagen hague hallid halliday harrington have here hierarchical hirschberg however html human ibdlidayan icslp identified including incoml incomplete information integrate integrating interface international intonation intonational investigated itltsurvey japanese johns kinds kluwer labeling labelling language lbne lete ling linguistic linguistics london longlilan look make management marimfi matches matthiessen mayer meaningful meanings might more motivation mouton natural next niques nuclear observation offers ohio ostendorf other oxford paper patterns penman petrelli phonetic phonetics phoneties phonological phonology pierrehulnbert pierrehumbert pinter pitch placed possible pragmatic predicted prediction presented press pretonic price proceedings processing production projects promising prosody ptvceedings publishers pwceedings realization references reflect remains researchers results reyelt rhythm same science sections seems sequences should silverman simple some speech spoken standard state step stress stresses study such survey syllable syllables synthesis synthesizer system systemic systems table taylor tbcusing tcch teaching technology teich text than that their them then there therefore thesis they this through tion tivoli tobi tolfi toll tone tones tonic transcription tsurvey turns typically under universify university used useful using uszkoreit version volmne well whether which widely wightman will with zaehen
http://acl.ldc.upenn.edu/C/C00/C00-2162.pdf	162	Improving SMT quality with morpho-syntactic analysis	acknowledgement alex algorithm alignment annual anslation arccs association athens august authors based brown canada categories christol christoph college comnmnity computational confercnce conference constraint contract corpora decoding della deutsche dialogs disambiguierung education empirical esprit estimation european eutrans evaluation ewfluation face fast fcrcncc federm ferencc ference fiir finland frainework franz fred gercg german greece gregor grmnmar haapalainen hans helsinki hermaml hermann hnproved html hzternational implementation international japan july june karlsson kobe language large leusch like lingsoft linguistics linguisties losef loser machine madrid majorin mariikka maryland mathematics mercer methods ministry mobih models montrdal morphologische natu natural niefien nodalida number pages parameter park parsing part partly pietra poster proceedings processing proeccdings project ptvecedings rcso references research roving rules running science search sonja spain statistical stephan summit support supported technology text thank this tilhnann tillmmm tool translalion translation under university verbmobil verl very vogel volume wahlster waibel wang weber with wolfgang work would
http://acl.ldc.upenn.edu/C/C00/C00-2126.pdf	127	Word	acquiring adam among analysis analyzer annual approach arouud association automatically better bunsetsus case ciples computational conclusion corl corpora cram defined della department depend dependency described discovers ella epartment estimates estimating experhnents froln iiiroshl iiozumi ilatzivassiloglou inforln informatics information institute james japan japanese japmlese keiryo kokugogaku kurohashi kuroshio kyore kyoto language ldnds likelihood linguistics list lnethod lnodel lnodifiee lnodifiers makol makoto maruyama mathematical maximum mectin meeting method model models morphological nagao national natural nihongo nlrl nsing ntropy ojun onvention order ordering pages paper pietra plus proceedings processing project recovery references remodifiers research ropriate rord sadao saekl same semantic shaw stephen structure survey syntactic syuei syupl syuppan takenobu tanaka target tendency tetsuo text third this tile tngmtge tokunaga tties ttural university used uses using valency various vasileios version vincent what which word yousetsu ytte
http://acl.ldc.upenn.edu/C/C00/C00-2093.pdf	94	P l a n n i n g t e x t s by constraint satisfaction	aaai across american amsterdam artificial aspects association automatic based belief building cambridge canada centering coherent colloquium computation comtmtational constraint csli customizing dale dechter discourse donnell doran dynamic editors embedded european experiments france from functional generating generation getting hentenryck hovy inmlligence institute intelligence italy iwnlg kibble knott lake languagc language levico linguistics logic maintenance mann manuals marcu markers mass mellish message natural ncai networks niagara nunberg oberlander organization pages plan planning power press proceedings production programming prvcecdings punctuation references relations rhetorical rosner satisfaction scott search souza stanford stede stochastic structure technical text texts theory thompson toulouse towards trees university using workshop zock
http://acl.ldc.upenn.edu/C/C00/C00-2166.pdf	166	Robust	about abstract abstraction adds agmentary agreement almost ambiguities analysed analyses analysis analyxes another approach approaches bank banks based been bianka bjgrn buschbeck calitbrnia capable carl center changing chicago coling comfutational coml comlflementizers comp computational conll constant constructing construction constructions content conterns converting coordination copenhagen corlms corpora correct correspondillg csli currently data database dealing deduction denmark described determiners dorna dust editors effort eichard einele eitm elliptical embody emnark encode english enhagen envisaged eport epresentation errors etween evaluation exclusively failures fere ferencc ference ferlmndo fiir first flat fledged flux formalism found from full function gambit general german given gives grammar heinecke hence hfformation hint hintikka however ictus immediately inconsistencies incorl inibrmation inlmt input institut intbrination intcrnatio interface interfaces internatio international into involved istiml iteaddriven ivan jaako johan journal julius karsten lang language large last lder lecture level lexical library lieated lieske linguistic linguistics lmda lmrxerx lnetho managing manii martin maschinelle michael missing mmmally module modules montague monthly montreal moravcsik more mori nationai natural notes ohan ohannes orated ordreeht orna ositional outlmt over overall package pages paper partial patrick pereira perform perspective phenomena phrase piece pinkal pollard predicate prepositions presented press proceedings processed prolog proper provides quantification queries real references reidel remaining rence representations resentations respects reyle rolmst rolmstly schiehlen sdfiehlen second selltencex semantic semantics semantikkonstruktion semmltic senmntic sets settled shieber sizable sprachverarl stanford sten step structure structures stuart study stuttgart subsequent such summary suppes system tational term that them these thesis they this throughput tlandling tlat together towards transfer transforming treatment tree truncated uage underspeeification ungary universitgt university urthermore used ustnexs usually utterances valuable verbmobil verbmol very vitx which with wolf words work works worm xpoken years yoshiki
http://acl.ldc.upenn.edu/C/C00/C00-2101.pdf	102	Learning Semantic-Level Information Extraction Rules by Type-Oriented ILP	aaai acquisition american analysis annual approach approaches artificial aseltine association attribute automatically based bottom built burth califf cardie classi combining communication computational computer computing conceptual conference congress conj crcnce crystal data diction dictionary down dumant engineering english examples extraction fication fisher foundations from generating generation hayashi huffman ieee ifip iiaruno ijcai ikehara inducing inductive information inheritance intciligcnce intelligence international itandbook iwanami japan japanese joint journal kaci kawaoka knowledge konvisser kurohashi language lcxicon learning lehnert lenert lexicon life linguistic linguistics ljca lktikci lloyd logic login machine massachusetts match mccarthy meaning meeting message methods meyer miyazaki mmming moldovan mooney muggleton nakaiwa nami nasr natural noun ogura oyama pattern patterns phrases podelski positive proc processing prog programming publishing references relational results riloff rules sakai sasaki semantic shirai society soderland springer statistical symbolic system taikei test text tkde tntcrnational transaction transactions translation translations typeoriented understanding university untagged valency wild with workshop world yokoo zelle
http://acl.ldc.upenn.edu/C/C00/C00-1002.pdf	2	Learning Word Clusters from Data Types	aaai abney acadenfic acquisition allegrini ambiguities ambiguity analogy annotated annual association athens august automatic based beil bell between bootstrapt boston bridge briseoe calzolari canada carroll centrc charles clarendon classes clustering clusters cognitive coling colinga computational concepts confcrcncc contextual controlled copenhagen corpora correlates cscarch deliverable dimensional disambiguation discovery dissertation distributional distrilmtional dizionario engineerin engineering england english epresentation evaluation experience explorations extraction fall federici fodor from garzanti gcncrativity grammatical granada greece grefenstette hierarchical inducing issue italian italiano june kluwer knowledge language lcxical lexical lexico lexicon light linguistics lrec manchester maryland mccarthy meeting methodology milano miller model montemagifi montemagni montreal natural notes nuovo object oxford oxtbrd pahner paradigmatic parsing patterns pedersen pereira phase pirrelli polysemy precccdings prescher press procccdings proceedings processes processing project publishers qnitivc references relatedness relations resolving retrieval riezler roach rooth schfitze science selection semantic semanticallt sense sentence series shallow similar similarity similariw sparkle special spring stanford subject symposium syntactic syntagmatic system tagging text thesaurus tishby transitions type umist university vector went where with word words working wzvng
http://acl.ldc.upenn.edu/C/C00/C00-1055.pdf	55	Toward a Scoring Function for Quality-Driven Machine Translation	adaptive adwait afifi aided amta analysis anlp annual applied approach approaches arpa atabase automatic available based bases beat bernhard boston boundaries bratko brian brill cambridge case chambers chandler chapman chinese christiane christopher clark clarkson cleveland cmucambridge collins computer connell daelemans data dekai density development dictionary doug dunn duxbury eacl eduard eibe electronic english entropy eric errordriven eurospeech evahmtion everitt evolution extraction fellbaum foundations frank future generative graham grammar graphical greece guide haines hall hans hatzivassigloglou hill hinrich hornik hovy html http ideutifying ilia implementations integrating ishwar ivan jakub japanese java jeffrey john jones kaufmann kcnkyusha kenji kenkyusha kevin kleiner knight knowledge kubat kurt langhorn language languages large lcxis learner learning lessons lexical lexicalised lexicon machine madrid manning masayo masuda matthew maximum mcgraw meeting memory methodologies methods michael michalski mining miornik miroslav mitchell modeling models montrdal montreal morgan multivariate natural nirenburg oxford pair parsing part paul philip port practical press proceedings processing quick rapid raskin ratnaparkhi reference references reynar rhodes richard rick riedwyl ronald ronl rosenfeld ryszard scale schutze sentence sleet software speech statistical statistics steve study systems tagging techniques three tilbm timbl tokyo toolkit tools tranlsation transformation translation tukey tuwien twisted universal university using vasilcios version virginia walter washington whimey white wiley william with witten wordnet xuanyin yamada york zavrel
http://acl.ldc.upenn.edu/C/C00/C00-1076.pdf	76	Extending a Formal and Computational Model of Rhetorical Structure Theory with Intentional Structures h la Grosz and Sidner	accotttlls alex analysis aniel asher attention barbara candace collabonttive compulalional computational dialogue discot discourse flmtorical functional grosz hilosophy ilding intentional intentions johanna karcn lascarides lbward level lheou lingttistics linguistics lingulvlics lochbaum marcu marm mart megan model moore moser multi need nicholas organization planning polhtck problem proceedings qmlational questions rages references rhetorical sandra sktner structtlre structure synthesis text theory thompson toward trees william
http://acl.ldc.upenn.edu/C/C00/C00-1003.pdf	3	Selectional R e s t r i c t i o n s in H P S G	acts allen alshawi androutsopoulos anications appelt aspect astructure batenlan benjamin cambridge canlbridge carpenter center chicago chical cmnmings computer core cornm database davis development driven editor editors enat engine engineering environment feature generality generation grammar grosz head infl information interface interfaces invest iticra jones kaufmaml knowledge kpml language large lczical lenat lezicon ling linking lmlguage logic martin merit morgan multilingual natural naturallanguage number ortability pages pereirm phrase pollard press ptvcessing readings references ritchie scale science semantics sparek stanford structure structures study systein technology tense thanisch theoretical thesis time transt typed understanding university webber
http://acl.ldc.upenn.edu/C/C00/C00-1042.pdf	42	Statistical Morphological Disambiguation for Agglutinative Languages	aduriz ages agging agglutina agglutinative alegria andreas annual application applied arning arriola arsi arto ased association ational aton atro augusl august austin barl bergen bmgu brill canada carat case categories ceeinl chapter chur church ciol cmal coling college colnbining coltmra coml comp compul computational confere constraint corn corpora corpus cutting description disambiguation disaml disammguation dismnl edin editors eech eeedings eemm endency enelope eric extended ezeiza fence flazer fung gramm gruyter gsei gsklmn gsktmn haji hand hankmner hasti hetirew hfllective hladks hods hological ieally igdat iguati iguation iiemal iinite ilepresentation ilker imri inde inguistie ining international irutational istics itai italy jorge juha julim july june kani karlsson kcmal kema kemw khan kupiec lagging lang langm language languages learning level lexical lexicon lezieal ling linguistics literary llazer lllll llollll lmyle logi lstrkish lvceedin madrid maryland meeting megyesi meth methods montreal morpho morphological morphology mort moshc mouton mpirieal natural nfercnce nguistie ning norway ofsl oint onstra onstraint ornan otlazer oving pages park parsing parts pascal pcndent pedersen pirieal practical predictiol press proach probabilities proceedin proceedings process processing proeeeding proj quebec rafl reebank references renec renee reputing rich ridge rive rocessin roeeedin roeeedinw roeeediv roving rsing rule rulei rules rural rvis sibun sigdat sign slbrmal smtccured sociatio speech srilm ssing state stochastic stolcke system tagger tagging teikkilii teling text tezt third tiir tila tilt tlaz tlazer toolkit trat trenl trieted turkish uage uisl uisties umeth unre unsul unsupervised untagged urizar utational uzzi vcessi vees veessin very veul vorksh voutilainen with workshop zhou
http://acl.ldc.upenn.edu/C/C00/C00-2173.pdf	173	X M L T r a n s : a J a v a - b a s e d X M L T r a n s f o r m a t i o n L a n g u a g e for Structured Data	alph alphawork also atml available avapro average bilingual bingham clark convert corn corp corporation data deach december dictionairies dictionary direction draft dssslsyn each ectx engine extensible file files formats formed from glazman group home htlp htmi html http iadc icgasys icopro indcx index input inputs jade james java jclark language legasys lexical libraries monoligual need november ofindcx omnima omnimark original output page pair patml preprocessed product project provide pure quccnsu qucis qunlnlal ransformation recommendation references requiring rule sgml sheets siml size softwarc sometimes source sssl stylesheet successfully syntax system tech technology there tiac total totalling transducer transformat transformations tree uscrs used users variety version well were will with working works xmltrans xslt
http://acl.ldc.upenn.edu/C/C00/C00-2154.pdf	154	W e b D I P L O M A T : A Web-Based Interactive Machine Translation System	addison addisonwesley alexander alleva altavista america americas amta analogy analysis anei anghorne anguage angust anslalion anslation apache appear applying approach arren arti artificial association avid babel babelfish based berners bert between bill broad brown carnegie chat christol christopher clare coijng coling comments component computational computer computers conference conmmnication consh corn corp cover coverage dbrence demon design dictated diplomat echnicm editors elithorn elke embedded english equcst example exceed executive experiments farwell ference fileno fish flanagan flstems foundation frederking free freetranslation gosling heorctical history hogan home hsiao html http hwang icial ihmng inachine informa input instant intcrnet intelligc intelligence interactive international internet issues ixlh james japanese jarkko java jirc journal july jvahtatiol kevin kurzweil lange langhorne langua language late lenzo lhlblications lingttistics llogan lnternatio lroup luage machine machines madrid management march mark martin mary mellon messenger messengertsm methodological microsoft modeling mslalion msntm nagao network nlechanical nslation oaiae october oikarinen ojflit osenfeld overview pages pangloss parser pccijication pennsylvania personal piritual powerfi press principle proccedirtg proceedings process project proposal protocol publishing qnitive ralf ramework rcderking real recognition reeder references relay rencc rmmation robert ronald rudnicky school science seligman september server service services singapore slcms software sophie speech sphinx spoken springerverlag statistical steele steven study symholic syntax system systi systllan telephone telliflence terry third tile time tion tluman toole translating translation transparent trcmslation tten tvat uclion udnicky ummit university user usma uwer vahmlion viking volume voss wcccdiltfls wesley winograd with workshop worldng wuen xuedong yang
http://acl.ldc.upenn.edu/C/C00/C00-2113.pdf	114	An	abeille agan anmtal approach argamon artificial base based bilingum cardie colin coling collocations computational conference gbnficricncic grammars haruno ikehara inticliigencc krymolowski language learning lexicalization lexicalized linguistics lnticrnational loshi machine meeting memory national natural noun pages patterns phrase pierce proceedings proceicdings procicedirtgs pruning references role schabes shallow tags translation using yamazaki yuvm
http://acl.ldc.upenn.edu/C/C00/C00-1010.pdf	10	An Empirical Evaluation of LFG-DOP	aaal academic according accuracy advances algorithm also amsterdam analysis anlp annotated annotating annotations approach architecture argument arpa arsing artificial assessment ataoriented auxiliary base based benjamins berg beyond bies biometrika black bloothooft bresnan cambridge canada canon chapter coling colloquium comparing conclusion constraint contribute contribution converted corpora corpus coverage csli current darpa data dealing developed differences disambiguation disjunctive dissertation dublin empirical english epresentation erivation established estimation estimators ettropean evaluated events experimental experiments exploit exploiting features ferguson ficient finally flickinger fonctionnelles formal fragments france frequencies from functional future geman generalized gerlnany germany given good goodman grammar grammars grammatical grove harwud have help higher homeceutre hpsg human hybrid icant illc improve inadequate indicate inside insignil intelligence interpretation into introduced ireland issue issues istributions john johnson journal kaplan katz kaufinann kluwer language learning leaving lexical lexicales lexicalized lfgdop likelihood linear linguistic lssttes macintyre marcinkiewicz marcus martin maryland mass maximize maxwell memorybased mental method methods minimal mitkov model modeling models montreal moreover morgan naacl natural netherlands neumann nicolov ninth number only optimized oriented other outperforms pacific paired palwing parameters parse parsing pdezler penn piwceedings plvcessing population pour predicate press previously probabilistic probability procedure proceedings processing provided publications publishers quantitatively recent references relations rennes report representation research resentations results riezler riven robust saarbrackcn saarbrticken satisfaction scha schasberger seattle semantic series shortest showed significant significantly sima special species speech state statistical statistically stochastic structure structures style successfully such suggests syntactic system technical technology teoretical test tests that theot these thesis this tomita traiuing transition treats tree treebank treedop treestructures tugwell turned unctional unification universit university unseen used using utrecht verbmobil virtual vohune want washington were wessel which with words workshop young
http://acl.ldc.upenn.edu/C/C00/C00-1060.pdf	60	A H y b r i d J a p a n e s e P a r s e r w i t h H a n d - c r a f t e d G r a m m a r and Statistics	abney adam adwait ages altese analysis andrew anes anese anguage approach archive arsing ased atistical augusl awdlable based berger bkmmga borthwiek brisawa briscoe briseoe carroll choieemaker coling colingacl colino coml computatio computation computational construct context corpora corpus coverage decision della dependency dept deternfinistic dictionary eacl echnical edition egrating electronic email empirical ence endeney entropy eport eprint estimator even filtering framework fujio generalized ginccrin goshiflmfi gramnmrs granmwms grannnar grannnars guag guide guido haruno help hitoshi hozumi hpsg html huff ibrisawa ichi ified ilistic information institute intcrnalionel isahara itiaxilnuln ititoshi japan japanese john jolm jounal kanayanm kenji kentaro kiyoaki kiyotaka kyoto langua languag language lawe lbch lbehnology lexical lexicalize linear linguistics liroshi llatnalmrkhi lnodel makoto masahiko masakazu masaki matsumot maximum methods minnen mitsuishi miyao model models mural nagao natural ndency nishida nlprb nolo obse occurrence ochasti october ojeck ooyama pages parser parsing pear pietra posterior practical prec probabilities probability proc procc proceedings processing references rence research ribut robabilistic robal roccssi roce rser rved sadao sahara satoshi science second sehabes sekim sekine shirai sixth sornlertbunwmich ssin state statistical statistics statistieal stephen stics stoclmsti structure stsde subeategorisation suishi sumoto syntactic takenobu takenolm tanaka technical text this thods tinite tinl tokunaga tokyo torisawa toshi transdue treeadjohfing trees tsujii uchimoto undersl unifieation university ural urnal urohashi using uter vahm vccss version very vincent virach wide with words workshop ymtji yuji yusuke yutaka yves
http://acl.ldc.upenn.edu/C/C00/C00-2160.pdf	160	Producing More tleadable Extracts by Revising Them	abstracting abstracts advances ahnost also although analy analysis annual appreciate autmnatic automatic based bloedorn bunshoukouzou cannot cases chapter coding cohesion columbia comlmter comparing conjunctions constructing construction context coverage cutting degrade degrades description discussion document does effects encouraging english enlarge evaluation even examples extract frequent fukumoto gates generation geralcl half hasan have hnprovement however impleinent implementation improve improved improving incorrect info inore input insistence intelligent international into iowa ipsj irttelligent japanese jasis jing juman kawahara klare kurohashi kuroshioshuppan literature longman main manage mani mann mathis matsumoto maybury measurement meeting minel mluce more morphological most nagao need needs nomoto nonmto notes nugier original paice parser pasting press proe proposal prospects protocols quality readability reason references reform results revised revising revision revisions rhetorical rules sealable seraphin shosou show since state structural structure subjects summarization sunmmry sununaries supplemented system techniques text texts than that their them theory thesis they third thompson thorough those through tiuman ttalliday univ university urohashi version where with workshop worse writer young youyakubun
http://acl.ldc.upenn.edu/C/C00/C00-1020.pdf	20	A Client/Server Architecture for Word Sense Disambiguation	aaai aimelet akroyd allc appear approach approaches arid arpa artificial atttomatic barbara based bauel building cambridge canada chanod church claude coling combinations computatimzal computational computer conference corpora corpus corrdard cotzference database datasets dependency dictiona dictionao dictionary dini disambiguate disambiguation disambiguator dismbiguating driven ecai edited electronic elisabeth english erence error european evaluating evaluation examplar example exercise extracted extraction fall fellbaum finite finland fiom first francisco french from gale ginger gold granada grundy hachette helsinki hnprovement human humanities information integrating intelligence international introduction isambiguation issue kaufman kilganiff kilgarriff knowledge langttage language large leaeock lexical liltguistics linguistics locolex lookup lxmguage madrid markup method methods mokhtar montreal morgan multiple mztctiotz natural networks neural object october optimised oxford papers parser perspective press probabilistic proceedil proceeding proceedings programs references report resnik resolution resourees ressources rivalizing rolls roux santa segond semantic semantics sense senses senseval september shallow siglex sources spain special standard state statistical stevenson stockhohn subject supervised symposium tagging technical technology text their tomaso tongue towell transducers translation university unsupervised using uter vdronis very verylalge volmne washington what wilks with word wordnet work worksho workshop xelda xrce yarowsky your zaenen
http://acl.ldc.upenn.edu/C/C00/C00-1070.pdf	70	Lexicalized	aaai aaal aation abilisl accuracy acoustic acoustics advanced advances algorithm analysis ansformation aral areas areme artificial ased assl automatic based been biometrika boston both brill building charniak chen ciass clmology component computer conference considering coral corlms data disambig dissert dissertation doctoral driven dtgging duda ecently eech effective eflbrts emnlp empirical english entrol eriments error estimation etlix extend extension fercnce fication formation forney francis frequencies frequency from fronl gging good gtnmmar hart harvard have hdelligencc hence hendrickson hmms houghton icassp ieee improved improvement increase institute intbrmation intclligence international into jacobson john joint kaist katz korea korean language languages large lccpol lcxieal lczicon learning lexical lexicalization lexicalized loss lvcessin made many massachusetts master maximum merialdo methods mfiform mfitbrmly mitigated mitltin model models more morpheme national natural nerally nonmliform noticeable orea oriental other parameters part parto pattern perkowitz population probabilistic probabilities proc processing proe prol ptveessing quations ratio ratnap reason recognizer reduction references relial rence requires resented resulting rkhi rocessing rpora scene science selective served sigdat signal simplified smoothing some sometimes spacing sparse sparseness spccch specialization species speech statistical statistics storage suffers system tagging teclmology terms text tficial that thesis this tion tively transactions transfbnn trying ules uniform unit unitbrmly university unknown usage used usefifl using vcessi very viterbi where which wiley with without word words xity
http://acl.ldc.upenn.edu/C/C00/C00-1048.pdf	48	A Rule Induction Approach to Modeling Regional Pronunciation Variation.	about accuracy adapt algorithms almotation analysis antwerp approach artifit aspektcn based berlin between bosch both brili brill case celex clarendon classification clearly complex compression computational concluding consonant conversion corpora covered crences cvicw daelemans daelomans data describing differences directions discovered ditl ditlbxences driven dutch eclmiques eech endent english ercentages error eterministic extract finite first flelnish flemish fonilex gillis given graphemeto heuven higher honeme honemie however iation icult igtree important indo induction intelligence international interpret intion issue istics kaufmann kstrukt language lazy lcavnin learn learned learning lenrning level ling linguisti linguistic linguistics llls llse mach machine magazine mateo method methods mish mlsformation mnguage more morgan morphological most mouton mskerk natural ncdcvlwn networks nnmely noted numerous oche onds order oriented oruyter osed overall pages papers paralm parser partof phonemic phonok pita plication press problem processing produced production programs progress prol pronounce pronunciation publisl qmnltitative qualitative quantital quinlan rarelysis redicting references regional relationshi remarks rent research resel resent results rnillg rommcb rommciation ronun rosenberg rresl rule rules schal sejnowski sign sitication some special speech springer state study studying suggested synthesis syst systems tagging tations techniques teehni terich text than that these this those tilburg tion told trained transcriptions transducers transformation transtbrmation transtbrnlationbased tues understandmfle using utch valuable vayiants verlag volunw vowel weijters were when with word workshol xford xicon york
http://acl.ldc.upenn.edu/C/C00/C00-2141.pdf	141	Local context templates for Chinese constituent boundary prediction	abney academic accuracy algorithm alld ambiguities anguage annotating applied approach arge assp assures attachment attlonmlically austin automalic backing based beijing berwick bigram bloothooft boundary bracket bracketing brooks cardie carol chae changning characteristics chincse chinese choi chunks church claire coidps collins columbus comlmter communication component computalional computers conclusions conference constituent context corpora corpus corpusbased coverage crence data david decision definition delector dependencies describe different easy efficiency emnlp empmcal english entropy estimation fast fiflh follows free from gramnmr guarantees high higher hinese hrase htlang huang identify ieee intelligent iournal issertalion jersey katz kluwer kshol language large learned lest level lexical linear linguistics lnduslrial local ltuallg made magerman malehing maximal maximum methods mode model models multiple natural noun nptool nrestricted ntethods observed ocation ohio only output overall overcome paper parser parsing part partial partially parts peking perspeelives phrase phrases possibility prediction prepositional press principle probabilistic probabilities problem proc procedure proceedings processing processings proeessing program projected proposed provides publishers ralph ratnaparkhi recogniser references report robert roceedings runs science second shallow simple slate somerset soog sparse sparseness speech stage statistical statistics steven stochastic suite summarized surpasses syntaclic syslcm tagging technical technology templates tenny texas text that third three through thus time training transactions trecbank tree trigram tsinghua types uang university unrestricted useful very votllilanmn weischedel workshop yarowsky young zhendong zhotl zhou
http://acl.ldc.upenn.edu/C/C00/C00-2121.pdf	122	Automatic Extraction of Semantic Relations from Specialized Corpora	aaai algorithms automatically based bigram brown chamiak class clustering collocations communication computational dellapietra desouza details distributional engineering extracting fall from fromraw generated gram grefenstette ilierarchies implementation improving itierarchical journal knowledge language learning liermann linguistics martin mcmahon mercer model models natural notes pelformance pereira phrase press principle processing references retrieving semantics series sextant similario smadja smith speech statistical symposium text tishby transitions trigram with word words working xtract
http://acl.ldc.upenn.edu/C/C00/C00-1036.pdf	36	XML and Multilingual Document Authoring: Convergent Trends	aarne action alfa amon analysis annual anslators applications appropriate artificial association authoring automata awfilablc awfilable based becoming between bibliopolis bols brun business chalmers clause coch comlmtatiom community comparing compose compt computational conclusion confe confelwnce conference connect connections conqmler conqmtational consideration constraint constraints content controlling current datatypes david deftnite dependent described document documentation draft driving ecembe editor engine english eransart evahmting exciting exlensible explicit february feedback femando fiom fiqucs first forest fornmlisms fornmlizing franca from genenttion generation grammar grammars gramntatical great growing hallgren have heoo here hmdlionistic hmguage hmguages home httman http huelligence huernational improving index inlg israel issue june language large lecture lhlguislics limitations lingua linguist linguistics little llallgren logic ltartley lzmguage machine magnusson maluszynski marktq martin meeting microsoft might mitzpe more multilingual multilinguat naples natural nordslrom notes novcmbe novel obtained ocument opportunity orest other oxford page pages paper parallel pardi paris part pereira point potential power practices prescod press proceedhtgs proceedings production programming proof proposals quality ranm ranta recommendation references related removing replacing representations research researchers rettce reusability rule rules schema science scott seems sgmi shorttut shown some special springel still stressed strltctttres structure support supporting syntactic tational tech technical telwatiomd ternational text texts their theorelical theory there these this thonms three tmique tools towards transformations translating translation under university used users using view warren when where which widely with work working xslt ymetman
http://acl.ldc.upenn.edu/C/C00/C00-2103.pdf	104	Granularity Effects in Tense Translation	academic account adverb adverbial adverbials after algorithm aligned alshawi among analysed andolph ansfer anti apply aptcr area aspect association attachment available based because been between bianka bilingual bras buschbeck calcul calculations cambridge canada cases chapter coling commotions computational configuration connections copenhagen core corpora corpus could cover creenbamn cross culprit data demnmk detail determined disambiguation discourse discussion dordreeht dorna douglas dublin dverbials eases editor eira emcle emphasis engine england english european example extend extent extract factors fcrcncc fernando from full future german germany given goal gralmlarity grammar granularity grmmhu guagc hans harlow have here hiyan holland ibingen iguities incorrect information interesting interfaces international interpreted into investigate involved ireland iscour jalmary johan june kamp language larger learning like linguistic linguistics lishers logic long longmml main managing martin meet michael miriam model modeltheor mond montreal moran motivation mporcllcs myrimn next nouns occurrence other overlapping pages particular parts paul period pertinent placed possible preposition present presented press proceedings quantifier quirk rcncc reasoning references rence resolutioii resolution reyle rupp sabatier schiehlen scope scoping second semantic semantics shall shown sidney some structures student summary temporal tense tenses that these thesis third time tomorrow toulouse towards training transfer translation translations uction universit unspecified until used values verbmobil wednesday week were when which wide with wolf work would wral
http://acl.ldc.upenn.edu/C/C00/C00-2092.pdf	93	Data-Oriented Translation Arjen Poutsma	abeill amhitectum amsterdam approach artificial beyond brown california ceedings cocke coling complexity computation computational copenhagen crmference csli della denmark disambiguation dissertation efficient enriching erence eriencebased experhnental ference grammar grammars helsinki hybrid illc institute intelligence international jelinek joshi journal krauwer lafferty language lecture lexicalized linguistics logic machine manchestel means mercer methods models natural notes number pages pelformance pietra probabilistic proceedings processing publications references robust roossin scha schabes series sima stanford statistical statistics stochastic substitution tags theoretical theory translation tree umist using volume with
http://acl.ldc.upenn.edu/C/C00/C00-1073.pdf	73	Automatic Optimization of Dialogue Management	aaai adapting aforementioned agent allows also alto annual appled application applications applying approach arise artificial association attribute audio automate automatic back based been better between biermann boves care choice choices coling companion compared complex composition computational conceivably confcwztcc conference confidence confirmation confl conjunction control could cremers danieli decisions demonstrate developed dialog dialoguc dialogue difference discourse discussion ditlhrs domains dutchtrain dynamic ease eckert ecognition eech eeeh eekere effect email empirical empirically ertise eustolner evaluating evaluation example experiments expert explore explored fabbrizio features finer fixed fllture from function generation gerbino grain graphies have here however human icslp ieee ilwestigate implemented improves incremental informative inibrmation initiative intclligcncc intelligence interaction interactive interesting international interpretation investigated kamm kearns kobayashi language larger learned learning lelnented levin linguistics literature litman lnethods long machine made measures meeting messages method methodology methods metrics model more much narayanan national niimi ninth nips njfun note novice novices ntccnth oken only optimal optimize optimizing oriented other over pages paper performance pieraccini plamfing planning plying poor potential practical practice predicted predicting presents previous proc processing proe programming project promer proposed rcncc reasking recognition references reinforccment reinforcement reliability respect respects results reward rewards sanderman scales scvcntccnth search searching several shows similar simulated singh space speech spoken spolccn spring standard state states stochastic strategies strategy study sturm subjective such sutton symposium systeln system systems systent systern task tclccommunications technology terminal test testing than that theory this through timetable traditional travel ttnnsactions tutorials understand understanding used user users utilized valker varied vccssin versions voice walker when which wish with work workshop would
http://acl.ldc.upenn.edu/C/C00/C00-2090.pdf	91	Multi-level Similar Segment Matching Algorithm for Translation Memories and Example-Based Machine Translation	about aclualitd affect algorithln algorithm algorithms aligning alizing allows ambiguitd ambiguity ambiguous amtlogies analysis anslation aupelf automatic based because bilingttal both brew bring cambridge canada center chosen chris church classical coling comment communication comparing compensate compulational computation computer concentrating contributions corpora course cranias difference different distances does each ebmt ection edinburg engineering english errors essential evaluation evolutionmy example fact figure final first fischer fisher fourier fron furuse gale garder generated gives grenoble have henry human influence information input integer japanese joseph journal journdes lack lafourcade language layers lcpage leaf lemmas lepage levels linguistic linguistics machine main memories memory milliseconds montrdal more mssm must natural only originalitd other over papageorgiou paradigms parallel parser pauenl performance pipcridis planas please possibly press problem program project provides randomly recognition refer references relative remit report represented research retrieval same scienlifiquc scientifiques segments sellers sentences shallow significantly similarity singapore sole solving specific speed string structure structures supplementary systems tatao tela text texteval that then theory these thesis this thompson tostring traduclique translation troisi university uref used wagner with without word words work zmslation
http://acl.ldc.upenn.edu/C/C00/C00-1078.pdf	78	Chart-Based Transfer Rule Application in Machine Translation	adam addition agenda ages ahmg aiipa aligmnent aligned alignment alignments amsterdam amta andrew annn annual ansb anslation antonio apply applying arametcr architecture arpa association automaticmly based benja best bilingum bobrow brown caraballo chart chitrao cntherine colh coling coliw company computational conduct confercncc conference contimmm corpora cost could crookston ctmrniak data decreases delb della detail detailed determine detm develoi dict discourse domain domimmce down ederking edge editors effectively eflfl engine eonlinlmin eonsideriug erimcnts eriments eriving error estimation eugene eurotra experiineni experime experiment feet figures first flltm forests from future goldwater gris grishman grishmml groui group have help hnical however ietra ilistie improving information international ionary ishimoto iyanslation jane john johnson kosak lang lawc lems less level like limited lines ling linguistics lower ltein ltistical lying machine machines macleod mahesh manum manumly many mark matching mathematics matsumoto mdow meaning mechanisms meeting ment mercer merit messages meyers microsoft middle miehiko migned minimum mins mlow montreal more moreno mstbr multi multilingual multra nagao naturally needed niak nirenlmrg nmslt nmsu nnlch noticeable noun nsfin nttack occurrence onor output pages pangloss paper papers parallel parse parsers parsing partee penn perhaps peter petr pietra pnrsing preliminary preserving pretbrence previous problem procedure proceedings produce progrmn prol prone prot proteus provi publishing quality quistics ralph rate rber recent recognizers references rencc report reported research rmation rmph robert rocccdim rule rules selectively sentence sentences sergei sgall sggvall shallow shared sharon shell shouhl since sixth size solving somewhere sone soup sparse speech statistical stephen structural successful system tbcused text texts that then these this through ting tinies tion tirst tisti tort transb transfer translation travels treetmnk typically typology uagc uistics unan using utsuro verb very vincent were which will with work workshop ymlgm zchine
http://acl.ldc.upenn.edu/C/C00/C00-2124.pdf	125	A p p l y i n g S y s t e m C o m b i n a t i o n to Base N o u n P h r a s e Identification	aaal accuracy acknowledgements adam agan algorithm also another antal appear approach architecture argam arsi arsing ased asso association base basenps beatrice berge best bosch brgetting building bused carlson cascades chunking classifier clausal close coati coling colingeo combination coml comlmtational complete computational computer constraints context coral corot corpora corpus cunfl daelemans data ddjean dehaspe dellapietra department disaml djean driven elaborate ellapietra endent english entrol entropy eovetical erformed eric erik ertbrmance etter etwork evaluate evaluating example exceptions finitestate five funded grammars grants gslli guide halteren hans harmflll have hervd hnical hnproving hrases http iation identification iguation iguities improved ination inators inaximum inclusion indet individual inibrmation instead interesting internal into jakub jomputational jong journal kauflnan kauflnann kets koeling krymolowsld lactic lance language large lead learn learner learning leaves lexical linguisti linguistics logic machine majority marcia marcinkiewicz marcus mary math maximum memory meth might mitchell mmotated modeling more morgan munoz nation natural nmthod noun ntal obust ombination orted other outt partial patterns penn peribrmed phrase poor port proach proceedings processing programming programs ptvcecdings publishers punyakanok quality quinlan ramshaw recognition references refinement rence report resolve result results rhttl rmance roach roceeding room rosen ross roth rovement roving rperimental rqfi sang santorini science seattle sensitive shalh shallow shows sing sloot smomo snow standard stei steven study system systemexternal systems tagging taking tates technical ternational tested text than that theory this thods through tilb timbl tjong tmblished tmii transformationl treebank uctivc uiuc uiucdcs unified unyakanok using utational vasin vccedings vceedings veeedings version vincent vote walter washington which with wordclass worksh workshop worlcsh worthwhile would yllv zavrel zimak
http://acl.ldc.upenn.edu/C/C00/C00-1083.pdf	83	Taking Account of the User's View in 3D Multimodal Instruction Dialogue	aaai adapted agent agents allow also altering amanda andr animated animations anles applied appropriately artificial bares based between brat brian callaway candace changes charles choose cognition coherent collaboration collagen commandtalk complex computational content context control conversation coordinated customized david decide described design detect dialogue dialogues documents dowding dynamically editor effective elisabcth elisabeth elizabeth enviromnent enviromnents esponding experimental explana explanations finkler from gawron generating generation going graphics grounding hans illustrated incrementally inethod instruction intbrmation integration intellifence intelligence intelligent interaction interfaces interfhce interpreting interruptive james jean jeff jiirgen john johnson keep language lester lewis manager mark match maybury mechanisms method mismatches modeling moore motor multimedia natural owen pages paper pedagogical perception plan planner planning point presented press procedual profitlieh proposed questions reality realizing realtime references resuming rich rickel rist robert rochester sequencing shows sidner software spatial spoken stent stone stuart subdialogues suit systeln system task that then theory these thesis this thmnas thomas topic towns track training traum university user uses vahlster view viewpoint virtual which william wolfgang
http://acl.ldc.upenn.edu/C/C00/C00-1017.pdf	17	Probabilistic Parsing and Psychological Plausibility	aaai aistic analysis anism arscr association atation atatiou bank based basis bcnjamins brmfl building cascadt case category cdings chafer chapter charni charter christot cmnl cognitive collins computational context corl corley corpora cover crock crocker cruz depend ding dings disnmbiguntion eatrice editors efficient ellce empirical english enth entropy eugene eutvpea flfilistic flmrgh frolli garrod grmmm hart hell hitcctu htgen husetts icial ificial ilisl incremental inem inquiry intelligence irtec isti ivlarmts john johnson journal ksfor language lczical left lexi lexical lfiil ling linguistic linguistics lisambiguation llta llvtl lmman lmst lnodels lntdlig london lown ltmge lvccssing martin mary mass matthew mcch mcchnnisms mech menlo merlo methods michael misms mmming mnlp mnrk models modulm montreal mrsing mtorini national natural ndatiou neies nguagc nicholas nition norway november nrchitectures occcdi oeessing ohlwater oling orner oxford pagc park parsign parsillg pcfg perform peru pickcring piekering pnrk press probabilisti proc proceedings processin processing providence psych psychology race ratio rational rcinkiewicz rcnce rcse references research ridge robabilistic rocc rocec rocecding rocecdings roct ronrk ross rybmd science sentence served sharon shop sing ssing ssociation statistical statisticm statistics steil stevenson stnl syntactic thorst thorsten time tions tirst tistical treet trkhi ttal tthew ttnat twait tximmn uafle university utational wide wvlc
http://acl.ldc.upenn.edu/C/C00/C00-2096.pdf	97	U n s c r a m b l i n g E n g l i s h w o r d order	about acceptable advantages against algorithmics alvcy ambiguities analyses analysis applied appriate approach attachment avoid awflfl based basis been built case cases categorial chart checking chicago clausal colchester comi compiled comple computation computational conclusions constant constraints constructed contains coral csli deal different directly discontinuous disl display driven each eancellillg edge edges editor effect empty engincering ensure eric essex extended extra extracted extraposition fairly five formal from fundamentals general generally generates generation global gralninar grammars greenwich guide have head house initial inquiry intraposition introduce introduction ious irly iteln items itself johnson kind laeed languagc language lecture leftand leftextraposition length lexically lexity ling linguistics london lowden mini modifiers more natural naturnl needed nfovmation nmmar nodes notes notions number obtain ollard order ositing overgeneration pages parsing pentiunl pertbrmalme phrases pollard positions practice press proceedings process processing processor proeessin proposed provide ramsay rather references related required retrieving right roecssiug rough routledge rovide rules rustin seconds semantics sentence sentences shifted shown sicstus simply since site small some sponsored steps stratford sufficient syntactic syntax system takes than that then theory they this time tmrformance trace treatlllellts turns twenty uhnan uistic unacceptable unattractive university unsaturated vents wasow when where which williams with wood word words workshop worst york
http://acl.ldc.upenn.edu/C/C00/C00-1022.pdf	22	Exogeneous and Endogeneous Approaches to Semantic Categorization of Unknown Technical Terms	acquisit aided annual antal ased assadi associatiol association ategorization ation autom berg bourigault cdiix cdinfls chapter chnical class clationshil clivernl clustering cmantic cnth colin coml comparativ computer corl daelemans distinguishing dztractim engincc ewduntion existing extending extraction first from fuji gories grnm guide habert hesmlrus hltegratc hods icml iguai information ings inguisl intc international introduction ioll ismnl ismnlfiguation issue itlfi iwnymna jacquemin knowledge lang large lczical learning lersen lexicon ling linguisl linguistics lying machin madrid matic meeting memory method mlknown models modifier morl mrus mtlfiguntion mult multcxt nantes noun occcdings oget ommd orpus ositiolfing owski pages platform proccc proce proceed proceedings program purational putational rchef references ring rnalional rqfc rrnc rsion scmanl sehx sens sense senses shop snik ssot state statistical study systems terminology text texts thes tilb tiltmrg timbl tion tiona tokun trained tture tzarenko uagc uation uilding university urmnoto using usscll utational utvpcan uval vdronis vktrvncc vrel word words wrccs wrces yang yarowski zweigcnt
http://acl.ldc.upenn.edu/C/C00/C00-1007.pdf	7	E x p l o i t i n g a P r o b a b i l i s t i c Hierarchical M o d e l for G e n e r a t i o n	acquisition adjoining ahnost alshawi american amsterdam annual anslation appeared approach aravind architecture association automatic bangalore benjamins canada cognitive computational consensus douglas editor ehud english ewfluation ferenee first generation grammar grammars group hierarchical hiyan htral institute international introduction isreal john joshi language lexicalized linguistics machine maine manasterramer mathematics meeting metrics mitzpe models montreal natural north owen pages parsing pennsylvania plausible proceedings psycholinguistically rainbow ramon references reiter relevance report research science seattle shona srinivas sriniw steve supertagging technical transduction tree university upenn whittaker workshop xtag
http://acl.ldc.upenn.edu/C/C00/C00-2122.pdf	123	L e a r n i n g to S e l e c t a G o o d T r a n s l a t i o n	aagc aalitics aation ackcd acov acts agmon alcxandersson algorithms alignmcnt ambiguity ammdi analyses andations anlp anslation application applications applying approach artificial artiticim asing atio ational august australia automatic based batlincr berlin better bundations buschbc canada ccognitio center cithingcr cmbcr coling combination combinatorical comlml comtml conf confcrcncc conference contc corn corncll corpora crll ctnai data demar descril description dfki diah dialogv dialoquc disambig dolmctschcn dorna dovetailing ecai editio eiscrson ejficicnt emcle empirical eurot extract finklcr fllkcdcrking france gallwitz garl generation german gmbh hmitz hnph htchinc icslp ihfl ijcai incq inellcs inequalities informatik innslation intelligence internai internationm interpreting iomfl ional isscrtation issues italy itcct itlm ivcl jnivcrsil joint journal kfissncr kiissncr kilger kipp kmaier knowledge koch languagc language large lcuven lhrms lhtjin licity linear linguistics liuschb logic logics ltction lyanslation lycnto machine madi maryland masch massachusetts mathem mattavelli mcth mcthodologicm method methods mmmcs models montrem most motzkin mult multi mzemclc natural ncrcmcntal ndatiou negotiation nicmmm nirenburg nonlinear nsth okcn optical optimization orlllcll orna ournm pages parscr partial pieccwise poke practical pragmatikbasicrtcs proc proccssiu proceedings processing project prosody protossing quantz ranslation rbmobil rcithinger rclazation rclazatiov rcprcs real reasoning references report rfat rlin rlthcorcticm rnal robust rsiti rsitst ruland saarbriickcn scgmantatio schich schmil schmitz schocnberg semantic siegel sigi sixth slation smfl spccch speech spilker spou springer statistical stcdc stics stockhohn strategy structure stuttgarl sweden sydlmy systcllts tancous tarico tation technis than tian tillmnnn time timiz tincar tion tionm towards tthcmati ulti understanding univ univcrsity universitiil university unplugged utations uudcrstanding vcrbmobih vcrlag verbmobil verbmol verl verlag very ving vits vohmw wahlsl wahlstcr wahlster weber wolf worksho workshop worm xmmersson yvette
http://acl.ldc.upenn.edu/C/C00/C00-1016.pdf	16	Binding Constraints as Instructions of Binding Machines	algorithm amtphoric anaphora approach aspects backofen barcelona based becker binding binditlg branco calder capstick chicago chierchia chinese chomsky coling computation computational constraint constramts constructiomll coreference course csli dependencies dini discourse disse distance doctoral dordrecht driven dylmmics eacl eagles erbach essli estival explamaion features final fong foris formalisms frank free from functional giorgi govermnejlt government grammar group head hpsg ijcai incremental indexation indexing ingria inheritance intra johnson kamp kaplan kathol klein kluwer koenig language lectures lexical linguistic lisbon lnquily lnquio logic long luxemburg manandhar marrafa maxwell mealling mechanism merlo mineur natural noord notes oepen opposition parsing perspective phrase pianesi pollard press principle processmg profit prolog pronominal reference references referential reflexives reflexivity reinhart report reuland reyle rule satta semantics sentential square stallard stanford structure tation templates theory ulliversal uncertainty univ university uszkoreit wccfli webelhuth with within working ymple ziji
http://acl.ldc.upenn.edu/C/C00/C00-1062.pdf	62	LFG	acquisition ages alto architecture association ation berkeley bresnan carnegie carol carpenter center chapter chicago comp computability cornp cprc crwm cscarch csli dahyml drivell driven dymetman ecidalfility editors epreseill eversibility final formalistn fornml generation gram grammar grammars grannnar grmmnars grosz guide hlternational http inherently inson interactive ional ipunctio ittslmrgh kapbm kaplan language lczical lementation lexical lexicalfunctional lieg ling linq liter llol logic ltcs maxwell mellon menlo mental natural nferc noord nowlcdgc ntat orbital orntal pages pale park pati patr pereira petal pollard press problem procccdings processing programming ptwccssing ranunatical rasc rclatio references remarks report research reversible ridge rijksuniversiteit rtittical semantic shieber sonm stanford stickel style system technical thesis tlai toad ttational tyson ublications ublin uistics unification university urational uropcan user uszkoreit utrecht vcecdings wedekind with workbe workshop writer xero xerox zaen
http://acl.ldc.upenn.edu/C/C00/C00-1056.pdf	56	An English to Korean Transliteration Model of Extended Markov Window	adaptive addison adva anscription approach april asian assoc association assp attd automatic bahl based bounds brown bward carnegie charniak choi class codes coling collection colnputer communication component compulational computational computer computerusable conference continuous current damper data december decisiolt decoding description developmem dictionao directional english epartment error estimatio estimation ettglish file fold foreign forrnation from generate graehl hangul holland htformation icassp ieee inforlnation information inguistics international iral japan jonathan june jung katz kevin knight korea korean ktset language languages lean likelihood linguistics lnfonnatioll london lucas machi machine machines management markov mathematics maximum mellon memoo merialdo method methods model modelling models mptotically mttdomfield multilingual natural naturallanguage networks neural nforlnation north oaford olutional optimum oriental orithm ormation pame parameter part phmtetic phonological phottetics pmbabilistic press probabilities proc proceeding proceedings processing reasoning recognizer references report retrieval roger rosen rules salton science seoul sharman slanfill slation speech statistical studies syllable syntactic system taggi tagging talking technical text tglish theory transactions transliteration transliterations transliteratiott trat trattslation trees tsukuba ttropy university various viterbi weech wesley with word workshop zgram
http://acl.ldc.upenn.edu/C/C00/C00-2086.pdf	87	Informed Parsing for Coordination with Combinatory Categorial Grammar	ackage actic agarwal ailed albeit altcrna ambiguity anab analyses analysis anes anlp applications appropriateness arsing ational based boggess cacegorial catcgorial cdman combinatory coml comput conceptions concluding confirnled conjmmtions conll constituent constructions consull contcztual contrast coordina coordinal coordinate coordinnte czts derived discourse each efficient egorial eisner elieve ends english ennsylvauia enterl eriments esolution fart faster generation genermly grammar half hilosophy hmtitieal inat inatory insuflment intonation journal jult kais kaist karl kiss komagata korean kurohashi language lapa lengths letection lexicalism linguistic linguistics lint llres llsed lnemory long love lsis marion matching morph mten muraki nagao nfof njbrmatio normal noun occurrence okumura ones onjun only park parsing parsiug pattern patterns pennsylvania phrase press prevost process prolog radicm rasc rcan reason references remarks resl respectively results retlect rise rllt roach roduee ruct semantic semantics semmlti sentence sentences sics similarity simple software specch specijyin spoken ssucs statis steedman stochastic structural structure structures successflfl symmetric synl synt syntactic table takes tbrm thai there these thesis through tiction tigures time tive tlmd tunnen type types under usefifl using vpertics werage with yang
http://acl.ldc.upenn.edu/C/C00/C00-1067.pdf	67	On U n d e r s p e c i f i e d P r o c e s s i n g of D y n a m i c S e m a n t i c s Alexander Keller, Joachim Niehren	about accomodation alexander algorithms ambiguities ambiguity amsterdam anaphoric appear axiomatizing based claire colloquium comp complexity computation computational computer conf consistency constraint constraints construction cornell csli david dealing deduction deemter deliverable denys dependency dept descriptions determining disambiguation discourse dnchier documentation dominance donald dordrecht duchier dyana dynamic editor eiyn ellipsis events filfite fleck fragment from gardent grmnmars hans hindle independence introduction iournal jeroen joachim johan joint july kees kluwer kmnp koller kurt kusclmrt lambdastrnctures language ling linguistics lncs logic logical loth manfred marcus margaret markus marriott martin mathematics meaning meet mehlhorn mfiller michael mitchell morv mozart muskens niehren operators order oroenendijk over owen pages parsing partial peter peters philosophy pinkal polynomial predicate press problem proceedings proe programruing progrmnming prveeedings radical rainbow ralf references repesentation representation reyle ruhrberg saarland schiehlen science semantic semantics shanker solving stmfley stokhof strnctnres stuckey susanna system talking theory thesis thomas tilburg time tobias treatment tree trees treinen tutorial umlerspecification underspccification underspecification underspecified underspeeification university unplugged using vccedings vijay weir with workshop
http://acl.ldc.upenn.edu/C/C00/C00-2155.pdf	155	An H P S G - t o - C F G Approximation of Japanese	abdel acknowledgments addition algorithms annual approach approximation asia association based berlin bernd carl carried carroll chicago complex computation computational confcrcncc conference contemporary context cuts daughters diagne dieter distributed driven during editor education efficient emcient experiments extend feature figure form formalisms framework free from gibbon grammar grammars grant grmnmar gruyter hans hansulrich head honorification hope hpsg hypotheses information interleaved international ivan iwpt japanese john jsrg kader kasper kicfcr kiefer kricger krieger language lexicon linguistics logical malouf meeting melanic melanin memo metzing ministry mouton natural over pacific pages parsing particles phrase pollard press proceedings processing pronoun ptvcecdings references report requirements research restriction restrictor results robust rule science shiebcr siegel some speech spilker spoken structure stuart studies supported syntactic system techniques technologies technology these this ulrich umch under university used useful using verbmobil walter webcr with word workshop zero
http://acl.ldc.upenn.edu/C/C00/C00-1077.pdf	77	Identifying Terms by their Family and Friends	acquire acquired aille akira algorithm also although analysis ananiadou andy anmfiadou applied approach association austria automatic automatically autoumtic based because beijing benefit between bourigault brill byance candidate caption chantal china clustering cncc cole colin coling combined computational concepts conclusions confc confct conferc conference congress context contextual contextum corpora corpus could creating cvalue dept described designed development disambiguating disambiguation discovcry distances domain easily ecmlse edition engineering england enguehard entrancing eric etween evahlation example existing experiments extended extensions extraction figure filrther firstly focuses frantzi from fuller gaussicr grammatical grefenstette groupings handled have health higher hlentifying htternationai human ideally identification identifying iiierarchical ilingual image important improves incorporating incremental independent informatics information innsbruck intbrmation integration interna international isth january journal justcson katz kind kluwer knowlc knowledge lang language large largely larger lauriston learning lexical library linguistic linguistics list lomains manchester maynam maynard measure measuring medical medicine medinfo mentioncd method metropolitan might mmlber monolingual more mtzi much multi multiword nantes national natn natural nelson nformation nlprs noun ntcrnational olson ontology order ossibili other over pacific pages pantera paper part particularly percentages performance phrases possible precision present proach proaches proc processing proecssinfl properties ptvccssing publishers quantitative quigley quisition rcncc recognition references related research resnik resource respect retrieval rminol roach rovement rplorations rule scope sdue secondly section semantic senses services sflue sherertz should shows similarity simple slmcitic small smeaton some sourccs specific speech statistical strength such surface switzerland symposium table tagger tailored tasks technical techniques temic term terminological terminology terms testing text that there thesaurus thesis this tics tile tional towards traditional true tune turtle types uistic umist umls university ural used usefifl ushioda using vahm valid value variation vatue very vnce whereby which with word wordnet words workshop world would wwlcdgc zurich
http://acl.ldc.upenn.edu/C/C00/C00-1015.pdf	15	You'll Take the High Road and I'll Take the Low Road: Using a Third Language to hnprove Bilingual Word Alignment Lars Borin Department of Linguistics, Uppsala University Box 527 SE-751 20 Uppsala, Sweden Lars.Borin@ling.uu.se Abstract	achievements ahrenbmg aims alignment always andersson anna annotator answer appear applied automatic best better borin cascades center cmpora comparison compulalioual compulatiomtl computational computatioual computer conference confereuce construction copenhagen corpolzt corpora corpus cowora department dept enhanciug equiwflence etap ewfluation filter froln general hein inducing information interactive ircs irical joint languages large lars lauguage lexicons linguistics link liuguis liuguislics llel magnus mehtmed melamed merkei merkel methods michel mikael models modular natural nodalida nordic pages parallel pcumlel pennsylwmia ping pivot plug presentation proceedings proceediugs processing project references report research roceediugs sagvall science sigi simard sping sprogteknologi status step tagging teborg technical text than three tics tiedemann tool translation translational trauslatiou undelwlauding uniform university uplug uppsaht uppsala vcessing word worksho worlds
http://acl.ldc.upenn.edu/C/C00/C00-1045.pdf	45	Pronominalization	achieving alternative american aravind back binding brennan centering chomsky clarendon cognitive coling complex correct discourse editors ellen english enrico expressions extending focussing foris frameworks government hcrc hustinx joint language lectures linda lmbrmation look loshi marilyn michael never noam ordrecht oxtbrd packaging pages press prince proceedings process proficient psychological reference references referential research resource sentences sign signers simons spontaneous strube structuring suri survey susan technical theory thesis vcesses vmlduvi vonk walker written
http://acl.ldc.upenn.edu/C/C00/C00-1021.pdf	21	Tagging of very large corpora: Tol)ic-Focus Articulation	according afraid also among amsterdam annotated applied articulation barbora based basis being both building cases clustering comparison conlplex constitutes content corpus cutbedded defined dependency dressier dynalnic dynamic dynanfic dynanlic dynmnic dynmnie editor ercrzce every experimental focus follows haji hajif hladkfi hlformation holding honour however ical iisovi inflective interface into jarnlila karolinum kluwer knows language less linear linguistics mary meaning more natural neither nlore node nodes ntanuscript only order ordering pages panevov part partee paul petr pfeiffer pfieek phonology point positions prague probabilistic processing prototyt quagc references research rule secondary semantic sentence sgall similarly since steedlnml structure structures studies subordinate subordinated syntactically syntax systemic tagger tbcalizers tbllows than that their then theoretical there thml thomas thus tile titan topic tree treebank tripartite under underlying unpublished valency vceedinfls washington well whole with word
http://acl.ldc.upenn.edu/C/C00/C00-2159.pdf	159	A Bootstrapping Method for Extracting Bilingual Text Pairs	academic across actiwties aitive also ambiguity applications aprroach attd atzd automatic ballerini based berry bilingual bootstrapping both building case centre change cognitive collection comp comparable compared complete computa computational computations computer computing conclusion conference conqmrable coocur corpora corpus cpartmcnt cross database davis dcerwester decomj designing development dictionary dictionaty document documents dumais dunning education eedinjg english error evaluation extracted extracting fiers figure flournoy folt forthation from fully furnas good grefenstette hfformation however hull ideal indexittg infomtation information informatmn inilail initail initial inlormation international irable irrelevant issue itze japanese journal kaufinann kluwer laboratory lancaster landaucr landauer lanfam language languages lantzua large latel latettt lation lcmguage learning lglications lingual linguistics littman llorlllal lnfbrmation lochbaum machine management masuichi mccs memoranda meta method mexico model models multi multilinettal multilingual news normal obtained osition oxford pails pair pairs papers paradigm parallel patent paws pederscn peters prcc prec precision proach proc proceedings proceedintzs processing pseudo punis quelying query recall references replaced research resohttion results retlcebased retrieval roved rshlnan scale schfinble schfitze schi science selection semantic sheridan show sigir singular special stanford state still streeter structure summit supercomputer table talc test tests text texts that thesaurus these tional tlicsis training translation translators tudics univcrsi university useful using vahte value with words worksh wterloo xliii zanettin
http://acl.ldc.upenn.edu/C/C00/C00-2130.pdf	131	Corpus-based Development and Evaluation of a System for Processing Definite Descriptions	able about according activity actually adequate adopted algorithms allowed also always analysis anchor anchors anies antecedent application arbitrarily arrived assigning attempted attempts automatic automatically available bach base based basis because been began best better between bridges bridging carter cases cause certain clark class classes classified classilication coded coml common commonsense competition complex composer compound comprehensive consequence consulted consulting corpus could crucial currently database decide decision definite delined denoted description descriptions details determine developed developing dictionaries difference different difficult discount discounts discourse discover discussed discussion domain done drilling entities entity equipped especially evaluation ewduation explicitly extensive fact feed feel fellbaum final first five follows found from furthermore general hand handle handling have head heuristic heuristics however hypernymy hyponimy idea identitied implementing implicitly indicate indicator industo information intended interesting introduced isbikawa itcrc kadane kind knew knowledge kuowledge last learn learning least lexical link lists literature living locations makes manual means mentioned meronimy meronymy method might mikheev modifying more most name named names needed nevertheless nominal nonhead note noun nouns object observed observing ollowing only order organizations overall packages participating pcrfommnce persons poesio possible precision previous prince process processing proper quinlan rather reader recall recent recognition recognizing refer references referring relate relation relations requires resolution resolve resolving restrict results right room search selling semantic semch sentences shallow shown sidnm siduei signflicant simple sister software some start stone strand strategy such supplied synonimy synonymy system systems table tant task text than that theflat their them themselves there therefore these this those three tile topic total training tree trying type types typical using usually varies version vieira well wells were what when where which whose with without wordnet words world
http://acl.ldc.upenn.edu/C/C00/C00-2094.pdf	95	Using a Probabilistic C l a s s - B a s e d Lexicon for Lexical A m b i g u i t y R e s o l u t i o n	accurate adam agan anlp annotated annual argument ashington assoc association based bell best bilingual bining cambridge carroll chosen class clustering computational compv constitutes constructed corlms corpora csnik david delibera detlef dimensions disambiguation disanfl dismnbiguation done dora embased emnlp empirical english enough equency equent evaluated examities experimental fits franz gleml glenn good granada hard head hybrid iation iezler iguation ilinrieh ilistic induced inducing induction information intbrmation itai itself joseph kilgarriff language large lesson lexica lexical lexicalized lexicon lezical limited linguaj linguistics look manner maryland mats meaning meeting method methods mmnnotated mnbiguity mnomlt models nlost noun once ooth osenzweig perfornmd philip preference prescher presented probabilistic probabilistie proceedings proves provided putational randomly real realistic references resolution restricted results rivaling roceediugs satisfim schfitze second selected selection selectional semantically semantics sense sensevaia simple slot stefan supervised tagging target targetword task tational tely that this tory tyom unsupervised upercomlnd using valence verb what which whose with word workshop world yarowsky yield
http://acl.ldc.upenn.edu/C/C00/C00-1041.pdf	41	Deletions and their reconstruction in tectogrammatical syntactic tagging of very large corpora	academia acknowledgement agency amtfhora amtotatimls amzotating annotated applicata applied approach appropriate argunwnt arpa articuhltion articulation aspects based bccn bpicfocus building bulletin burfifiovfi capture carried cases ccnpus colslplelllellts combination computational conference contemporao content coordinate coordination corlms crag czech dependency description differences diflerent dordrecht economical education eebank ellipsis english eomparisols evaluation example examples fifth fillmore first foctts focus fralllenel frames francisco functional generative glztmmar granada grant greenbaum haji hlavsa hmguage human international iqflective italiani iworkshop kaufmann kluwcr language large leech leorica lhocecdings linguislics linguistica linguistics llaii llaji london longman ltaji ltladkfi make marcinkicwicz marcinkiewicz marcus mathematical meaning means ministry missin modifying more morgan natural need nodes notes ordrccht other ovfi panevovfi paper part partee penn pragmatic prague predominantly presented probabilistic procedures proceedings processing project quirk ratislava really redicate references reidel reported research resources restored reusable rule santorini seem semalttic semantic sentence sentential sgall silem some spain structure structures studi suggest suplgorted supports svartvik tagger tagging tags technology tectogrammatical teli that them there these this through tile topic treatment treebank tromrtite verbal volume washington where while whole with within workshop
http://acl.ldc.upenn.edu/C/C00/C00-2158.pdf	158	MT and Topic-Based Techniques to Enhance Speech Recognition Systems for Professional Translators	abbot acoustics actual adequately advantages alternatively analysis approach approximately apps ation average based benefit cannot christie cnals cnet comparc compares computer conclusion conference constrain construct contenl cook corpus cost could coupling cwdualion decoder department described disadvantages document ecch effects efficient english entire eric error errors evah example extend extremely features first gift given gregg hmguage home html http incorporating index institute international into itochberg kcizer language large line llle loose ltochberg lvcsr magazine memory method methods model monterey more natural needed noway paper pcmag poor possible predict proceedings processing produced rate rates rather recent recognition recognize recognized recognizer reduces reducing references remarkable renals report results reviewed reviews richard ringger robinson robust rochester rovements science search second sentence sentences shows siegler signal simplicity small sophisticated source space speech stern studies suggest system systems table teclmical that then thesis this tiffs time topic train training translation translations translator understanding university used using vocabulary watch webb what which will word words would zdnet
http://acl.ldc.upenn.edu/C/C00/C00-2132.pdf	133	A Met;hod for Ac(:elerati lg CFG-l msing/)y [Js]ng ])ependency	algorithm annotation annual association based bigram coling collins constraints context data decision dependencies document earley efficient encoding example extended free fujio generative global graminars grammatical hashida http imaichi information initiative integrated japanese journal kauflnan language lexical lexicalized machine magerman matsumoto meeting method models morgan nagao natural orgs pages parc parser parsing pattern patternbased proc processin processing progress prospect prvc prvcessing readings references report schemata statistical stochastic structure syntactic system takeda technical text three translation tree using volume watanabe xerox
http://acl.ldc.upenn.edu/C/C00/C00-2128.pdf	129	A Statistical Approach to the Processing of M e t o n y m y	ablcx acceptability according acquisition ages akira amei anal anese anmysis annual appiica applying approach appropriateness ased association atsmnu automatic bachimont based been berland boguracy bouaud bran briscoc bruno bunpa case cditio chapter charniak chascn chicago christopher church class clues cognitive colinc coling computational cont copestake core corpora corporation corpus correctly csolut degree delia desouza dictionary difference domain enifer enjoy eric eugene example examples experimentally explicit extracted fass fcrrari fifty finding first five front generative george gincenl gram graph guistics hanks hearst helmreich hera heuristic hinrich hirano hitoshi hitsuji hiyu hyponyms ichiro icon ikai improve indirect information institute integrated intelligence interpretation interpretations interpreted ioruzl isalmra itcassessment iverson jacques jalmnese japancse japanese japmmse johnson katcit kazama kdzfi kelmeth kitauchi kurohashi kusumi lakoff lalstitute lames language lapo large lexc lexicography lexicology lgcel linguistics lisher literal live lmlguage lmsed lnel machine makoto marker marl masa masaki matsmnoto matthew mctonymy measure mercer meta metallel metaphor method metonymies metonymy mmlysis mmmal mmming model models morphological murata mutual nagao nara national natural ndations ninti nletonyllly nology norms noun nouns ntcn occssin pages paper parts patrick peter phors phrase phrases pierre pietra porar preach predicate press processin processing prolnising prwccssing publisher publishing quistics quite ranked recessing references related ropes rown sadao sakai schiitze science semantic semantics seventy shin shop statistical step stephen steplmne structure studies survey syntactically system takahiro takashi taph tatsuo technology terlll tested text textual that their this three tokyo towm translation travcrsal treatment ttiyu ulcd understanding university uomputatio using ustejovsky very volume wakao ward were what word work yamalmshi yamanashi yamashita yammnoto yasuyuki yori yoshitalm yuji zwcigenbaum
http://acl.ldc.upenn.edu/C/C00/C00-1040.pdf	40	An Integrated Model of Semantic and Conceptual Interpretation fl'Oln Dependency Structures	aaflc abstraction across ages allen allows alshawi also aman ambiguity american amia analytisch anatonfic anlbiguity annual applications applied aral arametrized arsing articular artificial assessment atterns automatic balhn base based bates bean bobrow brenncnstuhl brsker cambridge categorial centering challc chapter charniak chlflz clasp classes clearly clinical close closed cogsci coherence coling coml computational computcr computers conceptual concurrent conf confc conference constraints context contiguration coping core coufcrcncc coufcrcucc crea cution dalrymple data decisions defined dependency description design deutschcn deutsche deveh difti divided domain domains donm driven dynamics dynanfic eacl easy ecilications ecitications editor editors eijck einc emerging empmcal engine engineering english eouceptual eration eriem european fall finally fiunily formation from further general goldman grained gral grammar gramnmtical grant grmnmar grounding group hajieova hallges haudling hentata hierarchical hierarchy high hirst holland ieee ilahn ilic incremental inedical inferences information ingen inheritance inte integrating intelligence inter interface interl international interpretation interprotatiou intert into ioth itahn italm items journal kaplan knowlcd knowledge lanflua lang language large largo layer ldth lesmo level levels lexeme lexical lexicalized liahn lication lied liflhrent line linguistic linguistics llows lmsed logic logical lombardo macgregor maintenanco make mald mcchallisllis mcctinfl meaning meeting members milward mmck model moore mooting nagao narr national natural nden neches nfercncc nges nlal nldb nmchanism nmetional nmke nmkes north numl object omacker once onstraints orienco oriented origin ortcd orting orts other pages pago pereira perraris phenomena pollack pollard prec press principles proc proccssinfl proceedings processing provided ptvccedinfls ratho reference references referential related relations relationshil remains rent representation resentations resls retation reusing rindtlesch roduction roductious roles rolls romacker rovided rules scale schemata schenmta schmolze scnlantic search sehenmta seidonari semantic semantics senlantic senmntic separation simi since sneidernlan socou sohacht sondheimer spatial sprach static stract strict strube structure structures studies subsuml sulall symposium syntactic syntax systcm systems systenls tation tbrm technoh tepo term terms text thank that their these through thus tiahu tica tiil tion transactions tter type types umathematics underlying uniform unitication university untcrsuchunfl uses using uted vccedinfls vcrbcn verbcategorial verbwortschatzcs want wedekind weischedel well whether which with without woods word
http://acl.ldc.upenn.edu/C/C00/C00-1058.pdf	58	Automatic Thesaurus Generation through Multiple Filtering	abstracts academic accurate acknowledgement acquisition advances aids aizawa algorithm algorithms aligmnent alignment americas amsterdam analysis ananiadou anlp annual appear applications apply approach arcade ased association attcrn autoinatic automatic automating based benjanfins bilingual boston bouligmflt broad building chapter church clean clusters coincidence coling collocations coml computational computerm conf conference construction cooccurrence corpora correspondences cort coverage creat cross currently cyusyutu dagan daille darpa data deerwester dictionary discovery discrete distributed domain dumais dunning eacl earning edinbourgh eech eijk evaluate evaluation expansion explorations extraction fimded finch finding flesources fourth frantzi from fujisige fung furnas gale gaussier gawa generated generation grai graph grefenstette guessing harshman hatzivassiloglou highly hods hyougen idem identifying ifull indexing information instead intelligence ipsj itze jacquemin japan japanese jasis john kageura keyword kindaikagakusha kitamura kluwer kupiec kurohashi kyoto landauer langd lanning latent lexicm lexicon lexicons line lingual linguistics lmra location lookup luman mach management matching matsumoto mawaal measures meeting mekeown melamed method methods mikheev minimum monolingual mori morphological multilingual nagamochi nagao natural navigation nested next noisy noun ound parallel part patterns pedersen performance phrase planning point pora practical prec proach proc processing proe project projects promotion proper query quistics quwe recent reel references research retrieval rminology roach rules science semantic smadja society speech stag statistical statistics structure strzalkowski study surprise system systems taiouzuke taiyaku technical tekisuto term termight terminological terminology text texts textual tfiemstra thesaurus thesis this tifying tiomme titles tokyo towards transactions translating translation translations twcnte ubiquitous univ university used user using utilization veronis versiot view well word workshop yaku yonezawa yoru zoshinteki
http://acl.ldc.upenn.edu/C/C00/C00-2164.pdf	164	Perception, Concepts and Language: 7~oA~9 and IPaGe	acdc aclh agcnl agent agents ages agmn agreenm aktionsl allpea along anklin arch architecture argument articulation automation automatische berlin block brooks bsngle cambridge choosing chosen cole communication comput computcr coneel confused consensus control convctwation conversation cories corro csch cube datcn diate distribution done dureh editors egorize egory einen erzog eschreil esses eter evelt finkler forlns found frvm gcsprochcner gedata general generation given goecke gory grice have here hological hras hrase illl imvelt inflected infornmdon inkrcmcntcllen innings intelligent intention interm inth introduced istic ituman jackendotf jllsl journal kamstrut klotz langua language langv lausible layered lboid lelni lemma levelt lexemes lexi liil like likely ling listed llleu llll lnfix luare mainly mihle miiller milde mobil montagerot morgan morl morphological msser munl natural needed nerier notions noun nowlk nslm nszcncn nterfa ntsists object obotics often okui okul olridge onetics ossible oter pages paral peaking peters prec pred press process processing psycholinguistitally quader rcalzeitbcdingungen realized references reibungcn reiter rell ress ressed result richt ringer rlin robot robots rotmst selbstkorrektur sema semantic situated situationsund sprache springer stem stra strit subcat synl syntaz tctllal tcztgencricrung teilauloll these this tirst tree trigger ttlgthl ullllsllal ungen universitst untcr ures vohmm volume which wiirfel will wilt with word words workshop xmnl xolloltly yiehl york
http://acl.ldc.upenn.edu/C/C00/C00-2149.pdf	149	Context-Free Grammar Rewriting and the Transfer of Packed Linguistic Representations	addison addition algorithm along ambiguity ambiguous anette anslation approach argo association atgl august avcc based being believe best between billet bouillon chart chartparsing checked colference coling colline common compact computable computational conclusion constraint constraints construction context contextfree countability csli current description development dialog difference digital disjunctive dordrecht dymetman easily ecember editor efficient elements emele english enumerates eserving essential explicit face factorization fitce forests formal france frank free freeness french from functional gawron gazon generation gidlrrt grammar grammars graph graphs great green grenoble have hmette hybrid icai input interaction interface issues john jtme kaplan kent kluwer labs lang language lcmguage less lhlguistics light linguistic linguistics ltarrison lumi machine made madrid maintaining management martin masaru massive maxwell meal meeting method michael model montpellim montreal nagoya natural norvig oduction ofunderspecified only orna outputs over packed pages parallel parser parsing parts patmng permits phrasal prec presented proceedings process properties rayner reading references relationship representation representations represented rewriting ridge ronald satisfaction sciero seen semantics september shared shemtov singapore slightly source spoken stanford structure structures such summit system takes target technology than that theoty thesis they this tlllli tomita towards tracked transfer translating translation translator transparent under using variant vcrte verbmobih vert view voiro wesley whether which while with
http://acl.ldc.upenn.edu/C/C00/C00-1069.pdf	69	M u l t i l i n g u a l i t y in a T e x t G e n e r a t i o n For T h r e e S l a v i c L a n g u a g e s	ability ablex across additional agile also anlp applied approach artificial asked assessed authors based bateman been being benson boxes brief broader broadly building carried classified collected comparison computer conclusions construct construction content control cover created criticism cross czech dale demonstration description design directly discourse ecai engineering english errors etfective evaluated evaluation evaluators expert expressing extending final first flmctional from further generally generated generation given grammars grammatical greaves group groups gual hajie hdelligencc hladk iczicon ilingual improved independently inflective itself jcai journal judged kinds language languages lexicons like linguistic main manuals matthiessen methodology mnltilinguality more mult multilin multilingual native natural next nigel nmltilingual nominal offer onents originating overall period perspectives probabilistic proceedings progrmn prototype quality range received recommendations references reiter required resources ried robustness round rule russian semantic sharoff shown similar software source subjects system systemic systems tagger tcich technical text texts them then theretbre they this three tile towards training types ulgarian users valid variation volume well were which workshop zeng
http://acl.ldc.upenn.edu/C/C00/C00-2112.pdf	113	M a k i n g S e n s e of R e f e r e n c e t o t h e U n f a m i l i a r	able about already also amework approach attention based bauerle because been believe berlin call card carly cases centering change checking chicago claim clearly coherence computation computational conclusion considering consistency deal deals definite definiteness definites demonstrated denote denoting descriptions despite died different difficult discourse does dournal driven early editor editors elsevier erent erfect erring example experiments expressions extend fair familiar familiarity file firm first flmction following forthcoming framework from given good grammar grosz gruyter have having head helen helm here however illustrates implications indefinites inference intentions interesting interface international interpretation joshi kinds language lemented like linguistics local london long made make makes meaning meni menj mind model modeling models normally pages pertbrming phrase place points pollard practicality pragmatic pragmatics press prove proven question ramsay rather reasoning reconciled refer reference references referents referring regarded related require resolution ressions ries riety roach root russell schwarze science seem semantic semantics sense sets seville show shown sidner since single slept snow snowdrop start stechow structure subsets success such suggests system tackle taking than that their theo theoretical theoretie theories theory they this three tilburg treatment tunmr understanding unfamiliar unfamiliarity uniqnely university updating variety view virtue weinstein which wide with within workshop would
http://acl.ldc.upenn.edu/C/C00/C00-2097.pdf	98	Compiling	above across alshawi also analysis annual appear applied arpa attempting available averaged based basil battlefield blackwell bratt bresnan cambridge careful case caused cheyer clause clauses clear coinmandtalk combinatorial commandt commandtalk communications comparable compile complex complexity conclusions conference containing context continuing core coverage critical cross ctic data deal degradation dependencies developer devices dialogue dialogues difference directions dordrecht dowding drawback dutch egal encountered engine english error evaluation excluding expand expected extreme fashion fifth figure find first forreal found free frej from further gawron gazdar generalized given gorfl gramlnar grammar grammars guessed happens high hockey hope html http immediately increase increasingly incremental instead intend interesting interface into investigate james kapla klein languagc language languages larger linguistic linguistically linguistics lower manv marginally massachusetts mces meeting models modification months moore more most naturai natural next nuance number obvious obviously online only optimistic over pages peters philosophy phrase practical presented press problem problems proceedings processing programme project promise puhnan pullum rate rayner reasonable recognition references reidel related relative rels report results revealed robotic rqjects rules rvcecdings same savitch scripts seems serial serious significant simulations small solution speech speed spoken spring steady stent story structure subjects surprising symposium synta system technical that this though translator turn turning undisastrous unification unlinked utter utterances values variability vceedings version versions viavoice washington were what when whether which whose windows with word worst would year zaenen
http://acl.ldc.upenn.edu/C/C00/C00-2109.pdf	110	Backward Beam Search Algorithm for Dependency Analysis of Japanese	adwait analysis analyzer anese anlp annual artificial association atnaparkhi based berger cmculation coling collins comparison construct corpus criteria david decision decisiontree dependency deterministic divergence eacl ehara emnlp empirical entropy evaluation feature fujita generative grishman harry haruno heuristics hitoshi hozunli intelligence inui isahara iwpt japan japanese jowrnal katsuhiko kentaro kiyoaki kiyotaka kurohashi kyoto language lbkunaga lexical lexicalized likelihood linear liraitation magerman makoto masahiko masakazn matsumoto maximmn maximum meetin michael mininmm model models nagao natural observed only ooyama parser parsing parsiug practical printz probabilistic proceedings project ralph references resources roceedings sadao satoshi sekine selection sentences sharable shirai statistical statistics structm structure takenobu tanaka terminals ternational terumasa text three time tpanese trees trial uchimoto university using vceedings with workshop yoshifumi yuuji
http://acl.ldc.upenn.edu/C/C00/C00-2123.pdf	124	Word Re-ordering	aachen algorittuns alignment alignments anslation apparatus approach approaches april association audio august based berger brown coling college comparison complexity computational conf conference context corpora della dialogs document dynmnic emnlp empirical estimation face faceto ferent found germany gillett held hnprovcd http ieee ieken informat international iyanslation japan july june karp kehler kobe language large linguistics loint machine madrid maryland mathematics mercer method methods models monotone natural niessen nual number olleagues ordering parameter park patent pietra problems proc proccssing proceedings processing progrmnruing references rwtg saarbr sawaf search sequencing siam sigdat spain speech spoken states statistical summit technology tilhnann tilhnmm tilli tilltnann transactions translation under united university using verbmol very vogel wahlster word zubiaga
http://acl.ldc.upenn.edu/C/C00/C00-2170.pdf	170	Jurilinguistic Engineering in Cantonese Chinese: An N-gram-based Speech to Text Transcription System	accuracy accurate acknowledgement across additional administrative africa african agency aided algorithm ambiguity analyses applications asian asymptotically attained attd balance based behall belling between bounds caden cantonese case chan change character characters cheng chew china chinese code codes communities computational computer conduct confere conference construction conversion convolution corpus courtroon created delivers department dialects different domain domainspecific ecoding efforts enabled encoding encouragingly engineering enhance equality error fangan firm formula from further fuzhu govennnent gram guangzhou have held here hfformation hksar homophonous homophony hong honologically honous htformation iall iannao identify ieee improvement include institute internal international issues jelinek jinan judiciary jurilinguistic kaufmann kong language languages lauguages linguistic linguistics lntermttional logograt ltong machine made majority many mateo microsoft model modeling more morgan most multilingual multimedia multimodal natural november official optilnizing optimal organized other pacil paper phonologically plemenmry population possible presented press pretoria prinsloo probable problematic proceedings processing project provided ptwcessing publishers ransactions readings recognition references region reported research resolve self sels sequence services shallow shorthand size some south special specific speech statistical stenograph success suii support symposium synchronous system technology testing theog third towards training transcription trials trigram tuning types university uses viterbi waibel weights which will with yueyu zhan
http://acl.ldc.upenn.edu/C/C00/C00-1028.pdf	28	Explaining away ambiguity: Learning verb selectional preference with Bayesian networks*	almey aspcct away btternational cambridge chomsky cognitive database disamt easo editors ervi explaining fere fodor hiding hierarchy iguation ilall ings intelligent itow johnson journal kimthmn kntz laird lang lements lexical lexicography light line lint mental miller model models morgml odor ognitively pearl plausible prefi prentice press probabilistic proceedings proeessin references resnik rove ructure sciev sciousncss sele semantics semmiti sens serum straighttbrward strategy structure successful systems taggin text theory tiona told tttractive uagc uage university wards what with wordnet workshop worl would
http://acl.ldc.upenn.edu/C/C00/C00-1079.pdf	79	Representation and Recognition Method for Multi-Word Translation Units in Korean-to-Japanese MT System	afio ambiguio anslation approach arnold average balkan based blackwell ccpol coil coling collocation collocational comparisons component compound computational conference correctly each effectiveness english finland frequency gaps have high hurnphreys idiomatical idioms information international japanese journal korean lexical lexicon linden linguistics machine meijer method modalities morpheme multi mwtu mwtus needed nevertheless nouns only order patterns pohang proceedings proposed quial recognition recognize recognized references relatively representation resolution retriewe roaches sadler samples santos science semi seon sfer shows society speed sung table taek technology test thesis this tola tran transhttion translation treating units university verbal very were wessel while word words written yoon yung
http://acl.ldc.upenn.edu/C/C00/C00-2117.pdf	118	Text Genre Detection Using Common Word Frequencies	across american analysis annual anthorship applications arison association associaton attribution automatic avery baayen between biber bulletin burrows cambridge chapter classification coling coml computational computers computing conference confi constant corpora cross cutting detection dimensions discriminant distribution diversified eacl eisenbeis european fakotakis frequencies general genre genres health hohncs honore humanities infornaation inte international journal karlgren kessler kokkinakis language lexical lexington linguistic linguistics litera literaly literaw mass measures meeting metrics mormon narrative nexus nicely nunberg patterns perspective press pretativc proc procedures recognizing references register related ress richness royal schutze scripture series shapes sichel simple socieo some speech stamatatos statistical story studies style stylometric text texts theow twcedie university unless using variable variation vocabulary with word writing
http://acl.ldc.upenn.edu/C/C00/C00-2148.pdf	148	Integrating compositional semantics into a verb lexicon	aaai adele adjoining agent agents allbeck allnel altering amsterdaln amsterdam anlp annotation answer antonio anwind approach aravind ased aspects atial august austin autolnating autonomous badler baker based bases basic behaviors beijing beth bindiganawde bttilding california cambridge canada candito capturing ceedhtgs ceedings chicago china christine clare class classes coling colloquium combinatorial compnter compositional computational computer conference conq construction copenhagen copestake criteria dang denlnark depamnent department derivation description development dictionary doran dynamically eacl editol editor elljamins ence english escription explanatory expressions extensions ference filhnore fnnne forms fourth generalizations generative goldberg grammar grammars granllnar grannnar grannnars graph heuristic hierarchical httelligence hwestigating hwestigation icial instructions international intersective introduction iternation iwcs izim james january joilfing jolm joseph joshi july june kahane kalhneyer karin kipper kippm kluwer knowledge language laura levin lexical lexicalized lexicography lexicoll lexicon lexicons light linguistics lnlerlinxua lournal lowe ltag ltags machine madrid manaster marie martha maryland massachusetts mathematical mathematics matthew meaningtext moens montreal motion mposium multilingual national natural necessary netherlands nlprs nlueh nornlan onslruclions ontology pacific pages pahnei pahner palmer patrick pennsylvania philadelphia plalming plilmiples predicative preliminary press principle proceedhlgs proceedings proces pustejovsky ralna ramer ranslation refel references regular represent representation rgument rosenzweig saintl sanfilippo schuler science sehabes selltence selnantics semantic semantics senlantic senli sense seventh shanker siglex sing slrttclure spain spring stanford steedman stone sylmhronous sylwtin symposium telnporal tenlporal theory thesis third tilburg tile tlee toward trang translation tree ttrational underspecification underspecified units university using verb vijay voss washington wilh willialn willianl with workshoi worksholj workshop yves
http://acl.ldc.upenn.edu/C/C00/C00-2087.pdf	88	Interaction Grammars Guy Perrier LORIA, Universitd Nancy2 BP 239	adjoining advantages algorithms also appear base berlin blache capacity catdgorielles characterization class coherence complexity computational computer contraintcs coverage data database databases ddductifs decisive depend description does easily editor efficiently especially essential ette exhibited extend extract factorize fbrmal forlnal formalisms france from general generated gert goal grammaircs grammaires grammars grammatical grenoble grmnmaires grmnmar habilitation have iifformatiques information inria integrate languages langues lecture leeuwen level levels lexical lexicons linguis linguistic linguistics maintaining making massive matter model modularize notes ntially other pages paget panorama paris parsing programming properties propridtds rainbow references report research respect retors science semantic shanker smolka some springer success such syntactic systbmes technique than that thdories their these this thsse tile tiques today traiteinent tree trees unification universit update using usually verlag vijay wants weir when while whole will with
http://acl.ldc.upenn.edu/C/C00/C00-1049.pdf	49	Layout and Language: Integrating Spatial and Linguistic Knowledge for Layout Understanding Tasks	address analysis ande andreas annual applying artmeilt assigned association automated between blocks breaks ccur cdeb chung cohesion collm column computational computed computer contiint contimtation continuation converting cornell daniela date dengel document double eaks ended example expressed followed foriexample forra from graeme gram gramrfiatical higr hirst hwee ibbsitiv ifais ignre ility indicaee indicate indicator ithis jane japan jessica july june kieninger kristen layout learning lexical line linguistics lira mary maryland meeting model morris nagano name november ntial observettion occur pages paper paracnrapli paragraph point pote probabilities probability probal proceedings quistics recognize references relations report right science seglnentation segmetltation sentences shown sometimes space structure structuring summers system systems table tables technical teng text that then thesaural this those tile tion tlnd tohtml umber university using value where wherever white will word would yong zener
http://acl.ldc.upenn.edu/C/C00/C00-1071.pdf	71	LANGUAGES	addison adjoining against ages albert algorithms ambiguity amsterda analogies analogy analysis andr aravind ausmme bally basic books cambridge cbmputcr cbnccpts center charles christopher cmatical colnplexity common comp computer context contextual cours creative culy description douglas dowty elsewhere english enja ertdsz evidence fall ferdinand fluid franknrt freeness gdndralc gheorghe grammar grammars group haukioja hirschberg hoft iconicity ilie injbrmation internal itkonen john joshi journal kognitivc lang language languages linguistics linguistik linguistiquc lishing longest ltkonen luciml magazine marcus mart martin mass math mbara mctalinguistik mins monster much naomi natural payor peter philadelphia philosophy plications press problem processing provide publid qmatics reading reasonable references rehabilitation report required research rlos robert sager saussure science sechehaye sensitivity sept shieber solomon structural stuart subsequence syntax tadter technical tree tucs turku universa university ussi utational versus vide voca wandd wcndc wesley wissenschaflsthcoric york
http://acl.ldc.upenn.edu/C/C00/C00-2127.pdf	128	Toward the "At-a-glance" Summary: Phrase-representation Summarization Method	aaai abstracts acknowledgement addison algorithm althot analysis anlp anmzal annotation annual ased association author auto automatic autotnatic barzilay based beconte between bougraev bresnan brmation capability caption characterisation chicago chizanowski closed coling compttter concept conclusion consists contains content continue corpora doctmlent doctnnent document documents does domain driven editors effective efficacy ehara elhadad eport evaluation every expectation experiment experiments extracting fast fine firmin forlnal from fulfilled general generation give glalme glance global grammar grammatical gramnzatical have house hrase ieval infimnation information injbrmation inserts intelligent into introduced japanese jing kaplan kennedy keyrelation klein laboratory language lexicalfunctional like linguistic linguistically mani masuyama matic mckeown meeting members mental method methods mikami mitre miyauchi most naacl nagao nakagawa natural news nuage obrst ours participated phraserepresentation points pollard popular press proc processing program prototype pster realization reference references relalions relevant report representation results retrieval rett rksh roce rocessing rogram salience salton scalable sdair sentence sentences shirai showed shows sifting similar some special ssiny stlmmaw strltctm structures such suggestions sulnmarization summaries summarization summary sumnzarization sundheim sununarization system tagged tagging task technical technology text thank that they this thus timber tlasida trine ttead ttirschman tulomalic ueda university until utilize valuable wakao wesley will with words workshop would xociation yamazaki zechner zzlomatic
http://acl.ldc.upenn.edu/C/C00/C00-1053.pdf	53	Deixis and Conjunction in Multimodal Systems	abella algoritlnn also ammal analysis animations annual approach argued artificial associatiol association australia bangalore based below buntschuh callaway cambridge carpenter cohen coling communication complex complicating component computational computer conclusion conference conjunction constituent constraints construct cooperate coordinated coupled data deictic designing detailed determining development difabbrizio directory duncan enables england explanations expression feature fifteenth finite following functional generating gesture gorin gramlnar grammar grammars greatly grosz have help httelligence human hzteraction inany information inodeling integration interactive interface interpretation johnston jones kamm kaufinann language large lcslp lester level linguistics list logic longer madrid maps marcus mcgee meeting modular mohri more morgan multi multimodal multimodally narayanan national natural necessary nmltilnodal nmltimodal noun numeral occur other oviatt parsing patr performance permeates phenomena phrase pipeliued pittman possessives press proceedings processing proposed readings references riccardi scale schemata shaffcr sharp should simple simplifies single smith spain spatial specifically speech spoken state structure structures subcategorization subcategorize sydney syntactic taken temporal than that this tightly towns treatment typed understanding unification university utterance utterances vcessing verb verbal volume webber which whole wilpon wittenburg wright zeljkovic
http://acl.ldc.upenn.edu/C/C00/C00-1027.pdf	27	Empirical Estimates of Adaptation: The chance of Two Noriegas is closer to p/2 than p 2	access actually adapt adaptation analysis appeared automated available based berkeley better boston burstiness called cambridge church close colnpany context defined document documents empirical engineering english expansion floriau framework francis frequency full gale hearst hierarchical history houghton information introduced jelinek journal kucera laid language larger layoff limits lists methods mifflin mixtures modeling more much natural neighborhoods neighbors nonlocal other poisson press query ranking recognition references related retriewd sims slnaller speech statistical stop structure studying submitted such system term terms text than that then thesis topic topranking treated umemura usage weighting were which words worse yarowsky ynamic
http://acl.ldc.upenn.edu/C/C00/C00-1037.pdf	37	Formal Syntax and Semantics of Case Stacking Languages	absolutive account actor aens agreement amsterdam annius argmnent aspects assignment assmne australia berlin case centrum characterizing christian compose composition computational conference constructive context course david department dependent distance driven dynamic editor entries erode esslli following formalisms gives grammar groenink hand have head ical iluwa inorphology intbrmatica invariant iocated japanangka journal kangaroo kees kracht kraeht lacl left lexical linguistics lncs logic logical long mahmf marcus markinfl marl marlu material merging miehaelis mildly mystery nominals nonconfigurah nordlinger obert pages paper past pennsylvania philosoph pirli presented proceedings rachel references retord right rock semantic semantics semilinearity sensitive september sheet shoot shown side similarly smnantics springer stanford structure surface syntactic syntax ternational that theme thesis thus university variables veer verb vermeulen volulne weir wiskunde with without
http://acl.ldc.upenn.edu/C/C00/C00-1009.pdf	9	COMBINATION OF N-GRAMS AND STOCHASTIC CONTEXT-FREE GRAMMARS FOR LANGUAGE MODELING*	addition advances aerfai afia ages algorithm algorithms amaya analysi analysis annotated approach april audio austin australia bahl barco based been bellegarda benedi best between bilbao brack bracketed budal buikting canada casacuberta categories category ccogu chelba coling combining coml comp comparison composed computational computer cont context contextfl continuous convcrsatiomtl corpora corporm corpus correspond corresponds correst cted curves derivations described different distribution ectin editors eecl efficient elawarc english eslmfia estimate estimated estimation evaluation exploiting fercnce figure flgol flom free frmnework from function gamma generation gilet global graminatical grammar grammars grammatical grammm grams graphic grmmnars grmnmars guage have hungary hwiw iablc ieee ieei image imgc improved initial insi inside insideoutside intelligence international interpretation ithms itio ition itri jelinek journal laface lafferty langu langua language large lari learning lgorithm likelihood ling linguistics lltation lmges lower machine marcinkiewicz marcus maximmn means memos mercer meth ming mleva model modeling models montreal more mori multispan national ncdf nchez ndorf nedf obtained ochast ociatiou olnl onds ontcxtfl osenfe osto outside pages pami parameters part partially pattcrn pattern penn pereira perncr perplexity presented press probabilistic probability problem proc processed processing project proposed pwcccding rccogu recent recognition reestimation references relation represent results rning robability roscnfl sanfeliu santorini scfg scfgs schabes several sgmchcz shuchez sidney sing siolls smffeliu smnple solving spanish speech spoke spoken springerverlag ssing statistical stima stochastic street structure structures substring symposi symposium syntactic syntactical technology test tested texas this tion tisl tlmn toolkit torres trans treebank tributions trigrams udio understanding university upper used using utation utational viii villa vitcrl vlri vocabulary volume wall wang ward ween when which with word workshop young
http://acl.ldc.upenn.edu/C/C00/C00-2146.pdf	146	Structural disambiguation of morpho-syntactic for K o r e a n *	abney accuracy ambiguity analysis annual arsing assoc black categorial chart comparing computational computer connectionist coordination coyerage cruz darpa dependency domain driven dutch earl english ficiency flickenger gdaniec gennb grammar grammars grishman harrison hindle hoffman hyeok ingria interpretation ircs jelinek jong klavans korean language languages lexical liberman linfluisties linguistics magerman marcus mark meeting natural order oriental pages parsing pennsylwmia picky prec procedm proceedings processing quantitatively references relations report robustness rooth roukos santa santorini speech spoken steedman structural strzalkowski syntactic syntax thesis turkish university using voyager weir wonil word workshop
http://acl.ldc.upenn.edu/C/C00/C00-2116.pdf	117	Automatic Corpus-Based Thai Word Extraction with the C4.5 Learning Algorithm	adaptive amtual annual arrays association automatic based bell boundary california centre charoenpornsawat church coling collocations communicatiomj compare compounds computational computationallinguistics corpora corpus decision disambiguation document extraction feature ference fitwt frequency from gran hearst hwceeding ikehara japancse japanese jolu kawaoka kijsirikul language large learning linguistics machine magerman mark mated mathematical meeting meknavin method models morgan mort multilingual nagao natural ngram number open pacific pahner parsing phrases piwceeding proceeding proceedings processing programs publishers quinhm references report reseatrh robert segmentation sentence shannon shirai sixth sornlertlamvanich statistical statistics status substrings suffix symposium system tanaka technical term text thai theory tree uninterrupted using word words workshop yamamoto
http://acl.ldc.upenn.edu/C/C00/C00-2104.pdf	105	Experiments	aaai abney according acquisition adductive adjectival adjectives aestion ages amherst amit annual answer answering answers anti applications approach argument artificial association automatic bagiu barzilay bchnical berland bootstral brian brown bruce building callan caraballo cascades celex center centr chains charniak church classitication clustering colin collectkms college collocations comp computation computational concept conference construction coreference coreibrence corpora corpusbased croft database decision determining dietionarics distributional document donald dublin dutch elhadad ellen empir english esslli eugene even evidence extraction fernando finding finite frank from hatziw hehnut hfformation hierarchy hindle home htternational hylmrnym ical identification ilara ilotf indexiltg inference information infornmtion ings institute intbrmation intcunational intelligence intelligent international ireland issue itanks james jessica jing jones june kathleen kenneth labeled langua large lbwards learning level lexical lexicography lexicon lexicons lezical lezicoqraphy lhvceedinfls lillian line linfluisties linguistics linguisties lloark lournal maiorano maryland massachusetts matthew mckeown meaning meetin meeting meth methods michael millet morton multilevel mutual naffali natural nijmegen norms noun nounphrasae nouns november nteetin ntelligent occurrence page paragraph park parsin parsing part partial parts passage patrk pereira ping planck pmsie predicate probabilistic proceedin proceedings processing proecssin psycholinguisties ptvceedin pwcced pwcecdings pwceedings question rang rcnce references regina rencc rence renee report retrieval retrieving rilotf rntan robust scalable scales schmid second selflearning semantic semantk semiautomatic seventeenth sharon shepherd sigir singhal sixteenth smadja smmnarization special speciticity speech spotter ssiloglou state statistics steven structures strzalkowski systema tagging tational tech text texts thesaurus thomas tional tishby tomek track trees ummarization universal university using utational vasileios vcccdings vccedings vceedin vceedinfls very word wordnet words workshop xtract yufcng
http://acl.ldc.upenn.edu/C/C00/C00-1019.pdf	19	A u t o m a t e d Generalization of Translation E x a m p l e s	acquired adding ages agglomera algorithms algorithnl alion also american americas anghorne anguage anslatio anslation approach aralld arallel arse arzi ascd ased atlguag atterns august barrachina bilingual bootstrapl cago ccdin chesl chester chicago chin clhodolo clustering colv comlmtahonal compapp consorl could crawling ctor cvcnlh descril diclionarg dolinel ecember editors edua ella ellen england ennsylva esearch etrieval evin example exica fcrcncc fillers french from fully gaijin gical hcorel hdbrm hilil hine hird hnost hods html http hulguag human hzlcrnational hztcrlzational ianguistic ical icial ictionary ijmorclical ilar ility illguistic ingttistic interv into ional iotra issues juan july kgtfl lachi lalion langauge lazion lccl lcrna learn left lhdgaria lhlh lhml lical liminary lindi lingua llel llierarchical llovy lmplel lnlclligc lnvesciga machi machinc machine machmc mana mclhodoh mcthodoh mctl mercer methodological mexico miguel mining mmar mnting monolingual mtion natural nemnll neutral ngla ngland nglish njbrmation nmnt nowlcdge nowledge nsla nslation nual ociation ocke olla oorhees oossh oral orlium ossil pages pairs papers project qeme qlera rail ranslation rcncc references rencc rence rench resnil riven rmtslatio rnaiional roacll rocccding rocessessin rocessi roecedin roject roxvh rqlish rwei sant sergio slic slots snik sofia soup springer statistical ster strait strings such sues system teml than theoretical tically tion tive tony tracti translation treasury trujil tvncc university upenn urle usillg using utom vcalc vccedin vcecdings volun vvel which while wide without work world would xamp xlraclion
http://acl.ldc.upenn.edu/C/C00/C00-1081.pdf	81	A Stochastic	american annual applied approach artificial associated association based baum bernard bigram capacity ceedings charniak chelba church ciprian clustering collins computational conference confetvnce context dependencies dependency eeond elinek english ependency equalities estimation eugene eview extended finite frederic functions genera george gging grammar hdernational hypothesis improvement inequality information ings intclligence international itth john kenneth kernel language lexical lexiea limits linguislics linguistics lised magical makoto maximization mecti meeting merialdo michael miller minus model modeling models mori nagao national natural noun number ollazer pages parser parsing peeeh philosophical phrase plus probabilistie proceed proceedings process processing program psychological references renee rkov seven shinsuke society some state statistical statistics stochastic structure syntactic technique text three tire unrestricted using victor ward with word workshop xploiting yngve
http://acl.ldc.upenn.edu/C/C00/C00-2088.pdf	89	A Formal Semantics for Generating and Editing Plurals	academic accent acknowledgements addison agstuhl alexander amsterdam amsterdmn analysis anaphora anaphoric andman assigmnents associated baeucrle barwise benthem bilge brighton bulk cahill cambridge canada carried carrier ccmter clime coling colloquium complex computational conceptual conclusions conjunction context controlling coof cooper corefe covers csolution dale described describing determiner dinary discourse dordrecht ecai editing eemtcr eemter eijck eitct elsevim enee english enthem eration erivation erlin esprit esslli eterminers european evans events ewms expressions extension fighting figure fire firefighting fitted following foris formal france from funded furthermore generalized generated generating generation germany gmyter grass handbook have hnplicature ighton ilding impacts interpretation iscourse journal kamp keenan kibble kluwer kmhmel knowledge language lappin like linguistics link logic logical logically lurality lurals lynne many mass massachusetts meanings meant meulen mill modeltheorelic montreal natural neil networks nominal nominals noun object objects operations oriented oups papel paper parsons part particulm philosophy phrases piwek plural powcl powe power press proceedings proe proiect provided publishers pump pumps pumpz purp quantificational quantifiers quantiliers reading references referring reported representation research reyle roger schloss schwarze scope scott selection semantics sets solid sowa states stcchow stimulating stone structures studies such system systems temn terms text thank thanks them theory these this three tippe tipper toulouse treatment underlying university used version weak wesley westcrstfihl what which with within workshol workshop would yields york yott
http://acl.ldc.upenn.edu/C/C00/C00-1029.pdf	29	A Class-based Probabilistic a p p r o a c h to Structural Disambiguation	accurate additional adwait aelemans ages also ambiguities ambiguity annual approach appropriate association atnaparkhi attachment automatically avid backed based beijing bigram brill cambridge canada cannot case cases categories cccdings chapter chment chnology clark class clcction coling collins comparing computational conj conje corpora corpus cvcnlccnth cvcnth czical david dependencies dependency described dictionary disambiguation disanfl donald duxbury each eacl ectional empirical entropy eric estimating eugene examples fiflh fixed frames francesc frequencies further genera general generative hang harnia have hdbrma here hierarchy highly howell however human hzicrnalionol ibas iguation ijuropcan information instances integrate international into irelal iterative itself jakub jeff joint just knowledge kong language large lcarni learning level lexical lexicalised linguislics linguistics lion liri lizing look ltindle ltong lvcccdinqs madrid makoto massachusetts mats maximum mccti meeting melnory method methods michael model models montreal more mual naoki natural ompurational orhshop orteshop ositional over pages parldli parse parser parsing pennsylvania philip phrase pllilip populated potential preferences prel preposition prepositional press principle probabilities problem procccdings proccssi proceedings psychology rcncc rcsnik references relations relationships rencc resnik resolution resorts restrictions reynar roger root rooth roukos rule saliln select selection selectional semantic sense shown similar similarity sites slical smoothing solve solving spain sparsely statistical stephen stetina structural structure subcategorisation such superior tali technique tguistics that thesaurus thesis third this three through trained ublin university unsupervised used using vcccdi vcccdin vcccdings vcecdings very walter weir well when which will with word wordnet work workshop yarowsky
http://acl.ldc.upenn.edu/C/C00/C00-1006.pdf	6	T h e E f f e c t s of W o r d O r d e r a n d S e g m e n t a t i o n Retrieval Performance	ages aligmnent alignment analysis ancs ancsc aneso annual anslation apancsc applications ased auchi automatic baldwin based bilingual bols case chnical ciety coling comparison computational confcrc conjcrcncc corctical correction cranias croft crvised data document efficient eporl ering examl examlfl example examples expressions felkm fence fisher flom foundations from front fuji gauging gral gramms grant hiil honcin hrases ibqmgr ical ieice iiirano indexing information international ipsj jaist japan japanese jngine journal kawase kitamura lalmnc language lapanese large lasucs learning linguistics lmiqu lmtwcen lntc ltigh lulical machine manning manual match matching math matsmnoto mcctin mccting mcthodologic memory method mmflmr mmsc mort moudtolo mslat mtonmtic nagao naisquis naist nakamma nations natu natural natwral ngurtgc nirelflmrg notes omashnc orgiou oval pages pagt papageorgiou parallc pilmridis piper planas prenticehall press proc processing qical ract ransl rctmrt references research retrieval retrieving rimchin roblcm roccs roccssin rpcrime salton sato sigh sigir siinilarity similar smar smmmoto smnit statist statistical stevt string study supl system systo tanaka technique text texts thing tioual translation tricval trieval trmlslation trmmation tsutsumi unil university unsul uruse using uttascn uusupervised vccssing vclopmcnt version vohun volmne volume vorking vortc wagner wmalizing words yamashita yansaction
http://acl.ldc.upenn.edu/C/C00/C00-2098.pdf	99	A Context-Sensitive Model for Probabilistie LR Parsing of Spoken Language with Transformation-Based Postproeessing	abney academic acknowledgements acoustics adelaide also ammal anonymous applvach apportiotting architecture association attd based batliner berwick block boston boundaries brill briscoe cambridge carnegie carroll charniak chunks cmtference colleagues comments comptttational compulational computation computer cooperation coqms corpora cotcerence cuwing develolmteltt dialogs difficult dolnain education efficient ejfort enqfirical especially evaluation exact face federal focused fotwlaligation generalised generaliz generalization german grammajw grammar grant graphs htternational icassp ication icslp ieee improve improving information international iscoe japan july kiefer kiel kluwer kobe kompe krieger lallguage language lavie learning leartting like linguistic linguistics lmtguage lnui lxmguage makiug malouf mass match meeting mellon methods mfinchen ministry model most multi multiplicity natural niemann palwiltg palwing parser parsing patwing patwiug pennsylwmia philadelphia phrases pittsburgh pivbabilio pivceedings piveessing plvbabilistic press probabilistic proceedihgs proceedings processing processittg psycholillguistics ptvceedings ptvsodic ptwceediugs publishers rate references research reviewers rinco rlment robabilistie robust roceedings roviug ruland rupp schachtl schmid science should sidney sigdat signal single sornlertlamvanich speech spilker spoken spontaneous spontaneously statistical stent strategy structures stuttgart summit supported tanaka techniques technologies technology tenny thank that their thesis this through tiibingen tokunaga tomita translation ttelp under university useful using utterances valuable verbmobil wahlster weber with word work workshop worm woukl would wright wrigley
http://acl.ldc.upenn.edu/C/C00/C00-1052.pdf	52	C o m p a c t non-left-recursive grammars using the selective left-corner transform and factoring*	abney addison alfred almey annual arser artificial association atisl atist atistical automata ccord charniak cliffs clitli cmlfcrcnce coling coll collins comp compiling computation computational conference corner covington covj david demers deterministic down dtrt efficient englewood eorner eugene fence finite fourteenth fourth francisco generalized generative gertjan glewood grammars grmmnar hall head hltcdliflcnce hopcroft ieee ilall ilisl implenmnl ings introduction jeifrey jersey jllman john johnson journal kallfili kauflmmn kaufmann lallllllar lanfluagcs lanfluage languages lefl left leth lewis lexicalised linfluistics linguis linguistics linipdstics local lsycholinflui lttc mark marl mcctinfl mcefinfl mealles meeting memory menlo michael mnbiguities mode models morgan nationo natural noord nual oark orgall pages park parser parsing pereira philip prentice press principles probabilistic procccdinfls proceedings processin programmers programming prolo rallsforllls rcnce record references relating requirements rian rnando rocccdiltga rosenkrantz roximation scavch siga sigplan stanley state stephen steven stic strategies switchinq symposium tcrnational theory three tile translation ulhmm ulll unification using utational vesley vohmw volume wcccdings word
http://acl.ldc.upenn.edu/C/C00/C00-1084.pdf	84	A u t o m a t i c Semantic Sequence Extraction from Unrestricted N o n - T a g g e d Texts	according ages aistnara akira ammiadou analysis anese anishi anlp annual appeared approach argmnon association atistics august automatic autotactic available based bchnical bigram cactus call casual characterbased chasen chinese chnical church cient colin combine computational computer conclusion conference contain conventional corpora corpus dagml data dictionaries dictionary different domain drawn dynmnie ecoglfit efficient enables english example experiment experiments extract extracted extracting extraction exts fito fluation frequently fung future good gram have helps hfformation hnamfi hnanmra http iccpol icography ience iianks iliroki information institute international into issue itiroaki iuci japanese japmw jiang jimlg journal jsai junya keio kengo kenji kenneth kinds kita kitauchi krymolowski kumiko lage lang language languageindependent languages lapmlese lapmmse large lbrm learning lingual linky lint litnfl lmtguage makoto manual many march masal masalmzu master matsumoto meaningflfl meaningful meeting mefllod method methodology mishi mlese model mohisa mori morpholotical mory multi mutual nagao nakanishi nara natural naturm need nitta nmrphologieal nobesawa norms notate nouns number often omori only oriental osmnu other pages paper parser pascme patrick patterns phrases ppai precision preprocessing processing proper proposed propsed quistics recognition recognized recognizing references report representations retrieval ritten salvage sane sate segmenter segmenting semantically sentenc sequences several shallow shiho shinsuke shlomo simple simt slatisl smohisa sophia special specific state statistical statisticm statistics string strings such sysi system tagged tagging tags tatsuo tbilil technology term terms text texts that there thesis this tlirano tomo tomoaki tomohisa tools training translation tsutsmni tsutsumi undefined university uolin usages using vccssin ways were which whidl with witll word wording wordings words work worse written yamashit yoshital yuji yuwtl zagc
http://acl.ldc.upenn.edu/C/C00/C00-1031.pdf	31	An Empirical Investigation of the Relation Between Discourse Structure and Co-Reference	algorithm ammal amorrortu anaphora approaches aravind association attd attention august barbara based bolsfor canada candace case centering chinchor cognitive coherence coling computational conference constructing coreference corpora corpus daniel december definition different discourse discout distance eacl estibaliz experiments expressions extraction factors focus focusing framework functional global grosz herbert hirschman hitzeman hobbs hustinx hzternational information ings intentions interpretation janet jerry joshi july julyseptember june kameyama lappin large leass lettica lingua linguistics links local long lynette magdalena mann marcu maryland massimo matter meeting megumi mguage mitkov modeling montreal nancy october only operational organization pages perspective poesio practical proceedings processes pronominal pronominalization pronoun pronouns ptvceedings recognizing references referential resohttion resolution resolving rhetorical robust romera ruslan sandra scott shalom sidner simons sixth standards structure structuring study tagging task text that theory they things thompson toward trees university vonk weinstein wietske william wire workshop workshot
http://acl.ldc.upenn.edu/C/C00/C00-1008.pdf	8	Incremental	acquisition ames analysis annual approach apptvaehcs automatic barg bart based behavioral berkeley berlin bird brain brent briscoe brmalism bwards cahill cambridge carl carpenter case catcgorial ceedings chicago christopher clahsen cognitive coling comlmtational comp comparative component computational computations conference conjcrencc conncetionist constraint context coping covariation csli declarative defaults deliverable derivational deriwxtion description detmar dieter diisseldorf disambiguating disammguation disjmwtive dissertation doctoral dorna drivcn driven dsrre dyana eacl editor editors ehrlich entries erjavec erwerbs even ewan examples extension fclix feature from general gerald gerd german germanic grammar grammat guido guistics hahn handled hans harald have head hpsg incremental inflection infornmtion ingrid inheritance inhetqtance itcad ivan james jean jochen joehen john journal karen kehrein kilbury klein klemens klemmr knowledge koenig krieger language languages latter lczicon learning lexical lexicons lexieal lexikalischen lezieal linguistic linguistics linguistik ljubljana lwic magic manfi manuscril markus martin mass mbolic meta meurers michael minnen montreal more morphological morphology multidisciplinary must natural necessary nerbom networks niemeyer noun oazdar ognitive oint opladen pages paradigms part pects pedfeature perspective petra phology phonoi phonological pierre plural pollard press procecdin proceedings processing project publications rapaport rasc rase real reasoning reconsidered references relations renz representation representations repriiscntation richard rickheit riehemmm riloff robert robleni rovides rules rvzesse saarbriicken sascha scheler schnattinter science sciences simulation societal solution springer sproat statistical stitute structure study stuttgart subcategorization such susanne systems tational terminological text texts theory tiibingen toma treatment trost ture type ucturc ulrich unification universitft university unknown unpublished untagged until vceedings verlag vocalmlary walther wermter westdeutscher wiese william with wolfgang words wunderlich
http://acl.ldc.upenn.edu/C/C00/C00-1033.pdf	33	Authoring Multimedia Documents using WYSIWYM Editing	aaai abpi accepted alffl also amsterdam andr andrd applicalions architectures artificial association augustin authoring barker based belief believers bordegoni british classification coherent coling compendium computer computers conf conference constraint contexts declarative deemter deictic deixis demonstration directories docmnentation docmnents document documentation docunlent edinburgh edited employing enser esco esslli etai etrieval european evans evora exploration faces faconti feedback feiner ference ficial finkler first fluent francisco game generated generating generation geske geueration gmdstudien goebel graf graphics greeves hefley heterogeneous hffbrmation hfformation html http hummer iburth iconelass iconglass ijcai immi inclusion industry information inibrmation inte integration intelligence intelligent interactive interfaces interlaces international interthces isbn issue issues joint journal kaufmmm kevitt king language languages leaflets lecture leonegraphic levesque logic material maybury maytmry mnltimedia model montreal morgan multilingual multimedia multimodality natural neurohr news notes page pagination paraboni patient pharmaceutical pictorial picture plan portuguese power prec presentation presentations press principles problem proceedings processing procs profitlich programnfing progress publ readings reasoning reference references research retrieval review rijsbergen rist roth ruggieri school scott sequences sigir simonis smnmers solution space special springer standard standards stock systein system systems texts textual third tile tool towards trahanias unfamiliar user using vision visual vols waal wahlster wilson with workshop yellow
http://acl.ldc.upenn.edu/C/C00/C00-2102.pdf	103	Named Entity Chunking Techniques in Supervised Learning for Japanese Named Entity Recognition	accent ages alle almnese almse alnshaw anese anguage annual arning association atioils based borthwick chunking classification collins combining committee confcrcncc conj constructed contextual corpora ctnpora cucerzmt decision editor emning empirical endent entity esource evahmtion evahtation evideime french grishman hological inde international irex isvc jalmne japanese joint languagc language lanquaqc larg large lassit learning lexical lication lists machine maiorano matsui mcth mecl meeting message method methods minimally mmlyzer mmse mnbiguity models morphological multilingual named names natural ntity pages proc processing ptvccssin rcncc rcus recognition recognizer references rence resolution resources restora results rive rocccdings rstanding saito sassano sekine shinnou sigdat singer spanish speaker supervised task text texts tinding tion tipstei tree trmlsforma ttual unde unsupervised using utsuro vccssing vcecdings very work workshop yarowsky
http://acl.ldc.upenn.edu/C/C00/C00-1012.pdf	12	T h e effects of a n a l y s i n g c o h e s i o n o n d o c u m e n t s u m m a r i s a t i o n	advances amsterdam annotation aone aspects assochltion association automatic barzilay bird bjornar cambridge chains chinatsu cohesion computational computed context editors elhadad ellen forum function fundamental gorlinsky graeme hirst holland inderjeet indicator intelligent investigation james jane larsen lexical liberman linguistics linguistieum madrid mani mark mary maybury michael morris north okurowski organization pages phillips press proceedings references regina relations replacement robust scalable sentence spain sponsored steven structure summarizathm summarization system text thesaural using winter workshop workshw
http://acl.ldc.upenn.edu/C/C00/C00-1085.pdf	85	E s t i m a t i o n of S t o c h a s t i c A t t r i b u t e - V a l u e G r a m m a r s using an Informative S a m p l e	achieved ages ahnost also alternative although analysis annual anonymous applied approach approaches apptvach april argued atmey attribute attributevalue autolnatic based basis believe benefit better between bled blends both briscoe broad build built cambridge cannon capture carnegie carroll cdings chen combine comments complementary complex complexity complezity compression computational computer concentrate conference conjecture context cooet corpora could coverage cussens danger data dealing deals december della directly domfla editor effective efficient eirik ellees entrol entropy eomputationally erset estimated estimation etfects etliciently etter even example excessive experiments exponential extraction fall feature features fields fitting forests from gaussian gearailt geman given grammars grmmnm grmnmars hektoen here honald however ility inducing induction inforlnative informative inmtors instantiation intbrmative intclligcncc itowever james john johnson jorma june langua language larger latferty lcar lcarnin lead leaves lexicalised like line linguistics little llrl lmse logic long looking mach made malouf marl massach maximum maxinmin maxinmm meeting mellon miles model models more national natural nolo normalisation nquiry obtained ongoing open osborne osenfeld overfitting paclmd pages parse parsed parses parsing pattern pietra possible possil preferable prior proach probabilistic probably procccdi proceedings project pursuing qics rammars random reduce references relationshi relatively required research results reviewers rfms riezler rissanen ronald rosenfeld rule salnplillg sample sampling scale scenario scfg science scientific select selection semantic sentence sentences series sets should showed simple size slovenia small smaller smnple smoothing stanley statistical stephan stephen stochastic stronger stuart subcategorization such suggested supported symposi syntactic tcch technical template templates tested that there they this those tile tliank tractable trained training transactions tropy ttlan unification university unpacked unpacking upon usctts using valued vbabilistic viewed volmne washington ways when whole will with work world would zhiyi
http://acl.ldc.upenn.edu/C/C00/C00-2107.pdf	108	R e a l i z i n g E x p r e s s i o n s o f D o u b t in C o l l a b o r a t i v e D i a l o g u e *	about academic acts agent aistics allison also ation autonomous belief beliefs brian bridge cambridg cambridge carberry carroll cawsey chnicai clark cole colinu collaborative colnlmter comlmter communic communicative comnmnieation computational computer conclusion confl conflict consideration conversation conveyed copenhagen corpus correlate correspond could derived dialogue dialogues doubt doul editors eliefs eliet england eslitor eugenio evaluation even explicit explore exploring express expressions feature features forms future galliers gardenf generation grice gross harry have hmnan iation identified imlleratives instruci inter international into intonation ionso iterbert jennifer julia karen keith kern lambert langua language linden linguistics lingusitics lmboratory logan logic lones lranscrlpts lynn management meentrate model modeling money more morgan mputatio multi multiagent must national natm natural naturally negative negol occurring ormat other pages pape paul pemlsyi plan planning preliininary press proceedings protess provided radio rcncc rcnce realize realizing reasoning recognizing references refining report resented resolution response ression ressions retrieval revision role rose rules sandra saudra science show shows sparck speech streaking study subdialogues successflll synta system systems takiug talk tapes that theoretical these this tile ting tracts universii university uomputat used using utterance values vania vmlder will work your
http://acl.ldc.upenn.edu/C/C00/C00-1072.pdf	72	The A u t o m a t e d Acquisition of Topic Signatures for Text Summarization	aaai academic acquire admm advances against ally also analysis annual ansa applying aragraph arti articles artificial aseli aselin association ation atttonmtically august automated automati automatic base baselin baseline been beth brst bruc building case catei cessi chapter chin christopher chrzanowski church coincidence computa computational concept conclusion confl construction corporation cover crease croft ctiveness david dejong development dictionary diffe differen dings divide docum domains donminqmcitic dorization dunning ecai editor editors eduard edum eflk elcment eleventh ellen eneko enriching equal ercentage erformanc erforms erlbaum essential eurate european evaluation expal expansion extract extraction figure fimr final firmin fldl generating gerald girre grout hanks hinrich hmtu hnical house hovy http icial identifi identification iiorllls iladev illg iltlllt increase inderje inderjeer indicate information intelligence inter interpretation into inverse itearst ization jeffrey jinxi john journal kathh kenneth kevin kluwer knight knowledge language lares large lawrence lehnert length lengths lenqth lexicography lingui linguistics local lorenzen machine mani manning mark marti martin martinez maybury mdatious measul measur measure meeting method methods mfir micha mmarization mrly mtag mutual national natural nguistics nmlti olatz ompirical only ontling ontologies ontology ores other outperforin outt overview pagc pages passages patrick press proceedings process processing publishc query raction recognize references related relationships relevant rencc rence rent research resent resented retrieval rforman rformmlc rieval riloff ringle rocedure rocee role roughly rows saic schiitzc scores section segmenting senten shown sigir signatur signature signatures sllllllll smmnm sons statistical statistics steps steve strategies strzalkowski study subtopic summairist summarization sundheim surprise system table target terms test text texttiling tfidf tfidfmethods thdr their theory this thod thomas three tional tlilolf togetlmr tomek topic topics toplc translation ttirschman typical umrn under using vahms valuated values very vtic where wiley word work
http://acl.ldc.upenn.edu/C/C00/C00-2156.pdf	156	D e c i s i o n - T r e e b a s e d E r r o r C o r r e c t i o n for S t a t i s t i c a l P h r a s e B r e a k P r e d i c t i o n in K o r e a n *	aage alle also alum applied assigning aylor based bilisti bilistie break brie cmnl contl correction cotlt cquisil dctermilmd defi desir dirt dized eareflflly ecdi eech eeeh ehell ehite elll ells eorre eort erimenl erli error esis even ferc fiutherl gerl gills guessing gwon hirsehl houl hrase humli hybrid hyeok icatio iehard iencies ilill ilistic ilistie include inil ining into ions ision jong kore korean leeision leme lhml lict ling ltil lttcr luli lyze mach master megr metho mflnlmn mitalk mlknown more moreover morgan morphelne mtlers necess next nijmegen nmssive oeunbae olive ollll orienl orre osel oundaries peech phrase predic predict prediction prediel presenl press processi prot qllelices redi redietion references ress result results ridge rill rmamce rned roat roccedit roso rovi rpora rror rule sangho santen segmei shown simple sing sizth speech springerverlzg synergistic synthesis sysl syslxun systcnt system textto tgging thesis this time tioll tion tree trees trospecch ttexit ture uindm university urieux using veel well which will with working ylor yung
http://acl.ldc.upenn.edu/C/C00/C00-1043.pdf	43	Experiments with Open-Domain Textual Question Answering	aaai abduction ambiguities annual answer answering answers appelt approximations artificial ased association atabase attproximations based believe bigram bootstrat cccdings christlane class collins combination completely computational conference derived determismd discourse doug effect electronic ellen employs endencies ends erlbaum extraction fellbaum from functional generate goodrum have hfformation high ictionaries ignored iiobbs implemented inference information integration intelligence interpretation itobbs jerry jones justifications justifies knowledge lasso lawrence learning lehnert lencies level lexical lightweight linguistics llarabagiu lmrfol logical ltobbs malmscrlpt marius mark mectin mere methods michael mihalcea moldovan multi national nevertheless nlel nmnce operates overview pages paper parser pasta paul plng precise precision presented press proceedings processing produced project prover publishers question questions quistics references representations resentations richard riloff rocessillg rosie roxana sanda semantic shown similarly simple since sources statistical stickei surfing syntactic system tacitus taxenemies that thermore these this tile tool trec trove unpublished vasile very which with word wordnet yvendy
http://acl.ldc.upenn.edu/C/C00/C00-1025.pdf	25	Mining Tables from Large Scale HTML Texts	aaai acquisition along also ammal analysis annual another answering appelt applied assigning ation attribute because besides beyond building cell cells chained chapter chen chinese column computational computer conference conferetlce considered containing context corn corpus cues denotes description develop ding document documents documetttx douglas effects entity european evaluation even exalnple example expanded experiments explain expression extensions extraction fall ference fifth finch find fourth free from further future gaizauskas generation grammars green hlterj hltp hong html htternatiottal hurst hvceedings hyperlinks identifying improve index indispensable information informatiotl infornmtion interaction interpretation interpreting israel issue january keys kind know knowledge kong krishnanloorthy language languages lattguage layout learning like line linguistics lira lists litlguistics logical marker meaning meanings meeting message mikheev mining modeling monetary money multimodal named natural next note notes ontological oriental other package pages paths performance plain possible preliminary present previous price proccedings proceedil proceedings proceedittgs processing punctuation recognition recognize references relationship relative research retrieval retriew role saic same selnantics show sigparse simple spaces special still structure studied surrounding symposium syslems table tables tces technical text texts that their there this through tour traversal tsai tutorial undelwtanding undelwtatming useful usillg using usually value what which wilks will workbench workshop writers
http://acl.ldc.upenn.edu/C/C00/C00-2157.pdf	157	A Description Language for Syntactically A n n o t a t e d Corpora Esther KSnig and Wolfgang Lezius	advances andreas anlp annotated annotation annual applied athens august baden based beatrice bell bird brants brigitte building buneman cambridge canada carpenter carroll chiew cntation college computer conditional conference constraint constraints copestake coraputational corpora corpus csli definite deliverable denys description detlef deutschland docum dominance dorna dsrre duchier dyana edinburgh emele engine english essli esther estimation feature fibred finite fixed formalism franz free gabbay gang german germany gert glenn grammar graphs greece group guide hans hshfeld ihnenau infi inside internal international interpreted issue joachim jochen journal knowledge konvens krenn ksnig language languages large lexicalized lezius lilog linguistic linguistically linguistics logic london lrec march marcinkiewicz marcus markus martin mary maryland mats meeting mengel michael mitchell montreal natural newspaper niehren ochen october order outside over pcfg penn peter point prescher press proceedings processing programming proof query recent reference references relations report representation rewriting rfmi riezler rmation rooth saarlandes santorini scheme science search semantics skut smolka software solving special stanibrd stefan steven structures stuttgart syntactically syntax system systems tbrmat technical technology text theoretical theory thorsten tiger towards tracts treebank type typed universitiit university user uszkoreit version wang washington wfirttemberg with wojciech wolf wolfgang word workshop zajac
http://acl.ldc.upenn.edu/C/C00/C00-2111.pdf	112	On UNL as the future "html of the linguistic content" & the reuse of existing NLP components in UNL-related applications with the example of a UNL-French deconverter	accurate affairs algorithmics alternative ambe ambrunaz anglais annexes aofit april arabe ariane aslib association attalyseur attd atttomated authors authotw based basis blanc blanehon boitet brown capsogcti carnegie cash center champollion coling communication computational computer conf conference context contrat contributed controlled courant cranfield current dbmt ddbut deau developing development dialogue documents domains ducrot ectives edinburgh environment especially eurim europe existant fast fdvrier fischer foreign fort french gdudration geta grenoble guillaume have helsinki here high hltegrated hlternet human ification imag implemented information interactive interface international invited japanese japon juin july kant king lahting language languages last leader least lidia linguistics lingware london lugano machine maht mangeot many maruyama meisei melby mellon members metal methodology mind ministery mitamura mockup moneinme moniteut monolingual multilingual nantes natural networking nirenburg nouvelle nyberg oglno ohme other pacific pacling parser particular perez persl peterson petwonal pittsburg point practical prague press proc processing project projects quality rdalisation references report reported research revision romo rustin sakai science september sfatldard sitko sloeum smith speech star state stem symposium system taylor thbse third through tile titus today tokyo tomokiyo towards tpartir translation tutorial uchida university versailles watanabe ways wehrli well years york zelambrunaz
http://acl.ldc.upenn.edu/C/C00/C00-2137.pdf	137	M o r e a c c u r a t e t e s t s Ibr t h e s t a t i s t i c a l s i g n i f i c a n c e of r e s u l t differences *	addison address also analysis analyt applications aring assmnption assume assuml assumption atical avoids balanced being between chinchor church cmatical cohen coml commonly comp compared comparisons complicated computational compute computer conferc cornp corpora cscs dcolm dependencies differences difli ecial edition elnpirical empirical etween experiments forsythe hall harnett hirschman hmlter hods hunter iisl independence intelligence intensive introduction issue istical john jolm language large larsen lewis like linguistic linguistics llgl llletric llsed lnatched make many marx mercer message meth methods metrics midge moler more nathcm natural nettles next noreen often ples point possible precision prentice press processing publishing putc randonfization recall references rences results score show sign significance some sons stal statistical statistics stic such systems tational tbund test testing tests that they third this tificial tion topic ttction tural understanding understate used using vahmtillg values violated which wilcoxon wiley xyesley ypoth
http://acl.ldc.upenn.edu/C/C00/C00-2142.pdf	142	Rapid D e v e l o p m e n t of Translation Tools: A p p l i c a t i o n to Persian and Turkish	amtrup ankara april artificial berlin cognitive colnputing computer cruces development finite frmnework heidelberg hmnid incrcmcntal incremental intcrnational intelligence june karine laboratory language lecture lhlrkey machine mansouri mccs megerdoomian memoranda methods mexico multilevel natural nmnber notes overview pages persianenglish processing project rdmi references remi research science september shiraz singapore speech springer state summit systems translation trattslation university verlag workshop york zajac
http://acl.ldc.upenn.edu/C/C00/C00-2129.pdf	130	Multi-Topic M u l t i - D o c u m e n t Summarization	abstracts address amit among applications approach artificial azzam bagga baldwin barzilay based because breck chains chapter coling connectionist context coreferencing coretbrence could coverage cross designed doculnent document documents editor elhadad entitybased essentially fltsion gaizauskas generation generdtion gerard gild hasida hitoshi ihlmphreys inethod information intbrination intelligence into isahara ishizaki kathleen keinpen kevin langauge language linguistic linguistics martinus mckeown michael model multi multilingual natural nijhoff nmltilingual other pages press processing psychology rcncc references regina results robert saliha shop since space stage standard study summarization sunnnarization syun tags tagset tasks tbrmat text this transformation using utomatic vances various vector work
http://acl.ldc.upenn.edu/C/C00/C00-2161.pdf	161	Querying Temporal Databases Using Controlled Natural Language*	aalborg abiteboul about addison allen androutsopoulos appear applicatio applications approach artificial avignon barwise based benjamin benthem boehlen bridge cacm calculus california cambridge canada carpenter categorial chicago chomicki city clifford cliflbrd colloquial computational computer constructing contexts cooper copestake crouch cummings databascs database databases david department developin dordrecht driven edbt edinburgh editor editors elsevier emantics engineering esign etit exploring extending ference formal fornzal foundations framework france francez francisco gadia geert gelder generalization generalized grammar gramn grosz handboo hinrichs hnplemen http hull ificial igai igmod ilead implementation implementations intclligenee intelligence interface interfaces interval intervals introduction jajodia jones june kauflnani kaufmann klnwer kluwer knowledge kruijf language languages lectures linguistics lnformatik loflic logic logical maintaining manchester manuscript mateo meulen michael modality models montreal morgan morrill natural nelken norwell oceedings oehrle oriented pages perrault philosophy phrase phrases planning point pollard pragrustics pratt preposition prepositions press principled proceedings prover publishers puhmm qnantifiers quantifiers queries query querying questions ration redwood references relational report review richard ritchie safety science segev semantics series shrobe siga snodgrass sparck steedman steiner structut system systems tansel technical technology temporal temporality tense ternational thanisch their thema theorem theoretical theory thesis tics time timeg toman topor tracts transaction translation tsql type umcs university unpublished vianu wesley with zurich
http://acl.ldc.upenn.edu/C/C00/C00-2133.pdf	134	Prosody and the Resolution of Pronominal Anaphora	academic accenting acoust acoustically agresti algorithm allen alternative amlual ammal anafhora analysis anaphor anaphora anderson ansmission answer antecedent appages approach approaches arid association attd attention azzam back based book bosch brennan byron cahn cambridge career categorical cenlering centering chapter chicago choice ckert cognitive coherence colloquium computatiomd computational computationallinguistics comtmtational conq consortium constraints contexlt corpus course courtseling daarc data deictic deixis demonstrative descril dialog dialogs discomwe discou discoulwe discoume discourse duration editor editors eference effect effects eggert english erence ession esuppositions european factors fant fastl focus focusing framework frankfurt frelheinl frequency friedman functional gaizauskas gelling gentlenmn given graphical graphics grosz gussenhoven harvard heeman heuft highly humphreys ialog idiosyncratic ihaka illfluence improve ingtdstics intemisciplinao interpretation interviews istic john joshi journal kameymna kruckenberg laborato lang language lchiste limle linguistics listed ljngt ljnguislic lmkoff lntonatiolml local london look lund mass mcau mechanism meet meeting methode mmmication model modeling models nakatani natural never noante nonnp nputational ntax oken onmlelen orthen ostension oxford pages passo passonneau peaks perceptual perfornmnce perlbrmance perspective peter philosophy pitch pollard preliminaries preliminary press proaches proceedings processes processing progress prominence prominenzbasierte proninit pronorm pronouns pronsubj prontype properties prose prosodieanalyse prosody psychoacoustics qhhny quarterly question reader reading reference references referent referents relative relevance remote report repp reseamh resolulion resolution resolving rietveld roceedings rooth sandt sanlen schiffnlan second semantics signal significant signilicantly simple singer sjante speaker speakers speech spoken springer statistics status stem strair stressed strubc strube structure study style suprasegmentals swedish swulvik sydtology synthese table tasks tending terken tguage that theory thesis they this tion tire total trains tratns trod undanmntal universily university utational volume vowel webber weinstein wellformedness which wiley with xford york zttt zwicker
http://acl.ldc.upenn.edu/C/C00/C00-1046.pdf	46	Automatic Refinement of a POS Tagger Using a Reliable Parser and Plain Text Corpora	amano ambiguation analysis andrei annual anttila applied arto association astransac atro austin based brill brmation california case categow church compntational computational computer conference constraint context convention corpora crsonalization cruz department driven englis englisii eric errordriven ervised estricted extensions general grammar guessit haruno helsinki hielzhvhical hiroyasu htjbrmation intlwduction japan japanese juha julian kenneth kupicc language large lcarnit learning linguistics llidden llirakawa machine madrid march markov masahiko matsumoto meeting mikhecv mistalce mixtm model natural nogami noun number oriented parser part parts peljbrmance phrase pjvcexsing proceedings processing program publications references robust rule rules santa second selection shin shlq spain speech stochastic strings study summit syntactic system tagger tagging texas text third tlideki toward translatiojl trees tteikkilfi university unmq unsupervived using uxing very volume voutilainen washington word workshop yoshimura yuji yumiko
http://acl.ldc.upenn.edu/C/C00/C00-1004.pdf	4	E x t e n d e d M o d e l s and Tools for H i g h - p e r f o r m a n c e P a r t - o f - s p e e c h Tagger	ages aist analysis analyzer annual applic applied arning asahm association atabas automal based brill case cech chapter chasen colt computational conference contcxl cutting driven eport error european feat hological http ical ilistic imno inproveln institute japanese jatmlmsc jelinek jpllab july kitau kitauchi language learning length lermml lgrror linguistics lmicm lmology matsmnoto matsumoto meeting memory methods mistake mixtur mmse model morphologicm naist nara natural nfcv ofspeecll part pedersen pracl press probabilistic procccdings processing prolml puational rarchical recognition references riven rmatio roccssi rwcp schmid science selection sibun sigda singer speech statistical study tagger tagging technical third tinges tion tishby tlaruno transformation tsuda users utsm version with workshop ymmtshita
http://acl.ldc.upenn.edu/C/C00/C00-1039.pdf	39	B o o s t i n g Variant R e c o g n i t i o n w i t h Light S e m a n t i c s	acquiring agentifs agreement alternations analysis arampatzis asessing automatiquc bartning barzilay based binominales boston butterworth canada carletta case cdcile christiml classification coefficient cohen colingacl computational congr considers content context corpora cort diathesis dixidmc document each editor educational evalu evelyne extent extraction fabre find forces fran from fusion generalizations hadad henomena hrase identical ilar infof information inge inoininaux intcrprdtation interprdtatifs into jacquemin jean judith kappa kathleen klavans kluwer koster language lapata level lexical lexicon light linguist linguistic linguistics london mana maria maryland mation mawland mckeown measurement michael montrdal morphology morttreal multi natural nominal ormation ormational pages parat perspective phrase proccssin procecdings proceedings processing pronfising prvccedings qement qlais real really references regina reide results retrieval returns rijsbergen role romanistcs said scales scandinavcs sdquences sequences solutions sonm statistics strzalkowski study subjectifs summarization sychological synergy syntagmes syntax syst tasks term thesis this tomek tsoris types tzoukermmm universit universitd university user varia variant verbs view work
http://acl.ldc.upenn.edu/C/C00/C00-2175.pdf	175	Comparing	account accounted achieved also amount another approach ased assignment attachlnent automatic available based been better bosch brill briscoe buchholz building calzolari carroll cascaded center chapter chunk close coling comi compared computa computational conl consider considered corpus daelelnans daelemans daelenlans data desired determines difference different disambiguation discussion ditferences does drank eacl editor ehuuk eing emnlp empir english enough esslli example extracting fbrm federiei ferro final find finders found generalizations gfls gils gralnmatieat grefenstette grmmnatical guage guide have headword hell heuristics however http ical ilni important imrposes information instances itself keeps korbayovs kruijff large larqe lath lear learner learning linguistics lleport looking lose marcus mareinkiewiez melnory memory menlory methods mmotated mmotations montemagni more most national natural neighbor ning nlanner nmmory noun object only ooth original originally package pages paper papers partitioning passive penn phrasal phrase pirrelli point pora possibly prec prepositional produce produces question reconstructed reference references relation relations renee report representations resnik result results right roving rule rulebased rules sanfilippo santorini score scssion seeins seems sets sigdat significantly simple simt sing slightly sloot small some sparkle specification statistically stevenson student such syntactic systeln system technical test than that these this tilburg tile tilnbl tional training transformation transtbrlnatiol transtbrmation treebank tuber type used useflfl using utational veenstra verb verbs version versus very vilain ways what when which with work workshop wrote zavrel
http://acl.ldc.upenn.edu/C/C00/C00-2168.pdf	168	Acquisii:ion of a Language Computationall Model for NItA"	accusative active afferty affinnative after agrees albany amta angnl anguagcs animate annual approach approval association atltonlatic atttonaorpholo automatically based berkeley binding boas brill brown calit cambridge carning case checked china chms chomsky claltse class clause closed cocke combined completeness computational conclusion corpora coverage deducing delhi della demandside density dependency depth development direct dordrecht ecturcs editing elicitation evelopmcnt every facuhy final foris form from gohtsnlith goldsmith government hammar hand hawilla hology hotactics http humanitics icxical ihce ihmi implemented inchtdes inflectional ingttistics introduction itumimcite java jelinek jing jones language large lcxjcolls linguistic linguistics machine magerman main marcus mative mean means meeting mercer minimalist morphology natltral natnral next nircnbt nominative november number object opcn order pacific pahner pair papcr parameter person pietra practice prelbrred present presented press proceedings processed processing program purl rammar rapid readable realized references restricting result results rifles right roossin rttssian rule rules russian same santorini scllli scrgci scripts semantic semantics sentence shankcr side stage state statistical statistics structure subject supply support symposium syntax tbllowing tense that thcadth theory these this tile towards translation translations ttser twisted uchicago university unsupervised user using values verb vijay voice when with word workshop woval york
http://acl.ldc.upenn.edu/C/C00/C00-1047.pdf	47	A Method of Measuring Term Representativeness Baseline Method Using Co-occurrence Distribution -	accurate ammal analysix analyzing ananiadou appendix assigncd association asusume automated automatic betweenthe bilingual boolean brmation butterworths calculation caraballo charniak church classes clse coincidence collection collocation collocations colnparativc compound computational computernt conq corpora correlation cost criteria daille determining development distribution document documentation dualna dunning emnlp environment evaluation expressions extracting extraction fiequency fioln fiom firth fnlprs frantzi fronl gaussiel generalion given graph hisamitsu hlformation hlternational hnportant index indexing intbrmation interface ipsj itanks item items iwayama japanese jones journal kageura kando kate kita koll lange language lbpic lexicography linguistic linguistics lkeda lnlbrmation london lwayama matsumoto mcgill measure measures measuring meeting method methods miyahara mizutani monolingual mori mutual nacsis nagao nakagawa natural navigation nested nishioka niwa norcault norms noun nouns nozue ntcir occurrence ocuments oddey oflral ojipsj otomo oxford pertbrmance philological prec processing query rank ranked rankings recognition references report representation representativeness research resemz retrieval review salton schemes scientific sigir signl similarity society sparck spearman specification specificity statistics storage sttiii studies study surprise synopsis takano tcramoto technical telill term terminological terminology terms test text that then theory topic towards trans tsujii umino uriyanaa utational utseof values weight weighting where wiss word words workshop yang yano
http://acl.ldc.upenn.edu/C/C00/C00-1014.pdf	14	Reusing an ontology to generate numeral classifiers	acknowledgments aimai akio almual amano ambiguous amsterdam analysing analysis analyzing anchoring anmtal anonymous approach asahioka assignment association august authors backhouse baldwin based benjamins bond canada carnegie case category cdrom christine class classification classifier classifiers classifying coling comments computational conference copenhagen coqference corpus correspondence daniela database downing editor electronic english expressions fellbamn first floating francis fukumu general generative hayashi helpful hideki hirakawa hiromi html http hzternational ichiro ikehara international iwanami james japan japanese japaneseto japatwse jolm july kaisekih kamei kazunori kbmt kentaro knowledge korean kurz kyonghee language lanl lexical lexicon like linguistics lzmguage machine maintenance masahiro meeting meisei meishiku meknavin mellon method mitsuaki miyazaki montreal muraki nakaiwa natural nirenburg nlprs notes noun numeral numerical nunwral ochiai ogura ooyama oxford pacific pacling pactific pages paik pamela pantachat phrases press processing proiect pusteiovsky quantifiers references reviewers role satom satoru satoshi semantic sergei sfiry shimojo shin shirai shoichi shoten sornlertlamvanich surapant symposium system systems taikei takeru thank their tier timothy tokyo translation troduction university virach volume volumes wantanee with wordnet yokoo yokoyalna yoshifumi yoshihiko yoshimi zmguage
http://acl.ldc.upenn.edu/C/C00/C00-2106.pdf	107	Parsing S c h e m a t a for G r a m m a r s w i t h Variable N u m b e r and Order of C o n s t i t u e n t s	aagc about acta addison ages alget algorithm algorithmic algorithms amil amsterdaiil analytic annual arsing association augustin automa automata berlin bibliotheek blackl blcaux bresnan brown calnbridge chapter chnologics city clair clation clmls communication computation computational computer constraints construction context cover croft cynthia david deduction direct earley editor editors efli eprcsentation european extended fcrenee fernando filed flmctiollal formal formalization frank free gazdar gegevens generalized geoffrey gerald gramlnatical grammar grammars grammatical gramntar haag heoretical herzog ibrmal icnt iiot ikformatica ilosophy infix into iowa iwpt james jetfrcy joan john kaplan karl klaas klein koninklijke kwell language languages laug lbch leermakers lexical lilog linguistics lnai lnbvduetion meetin meeting mental methods meyer michael morawietz murray neil oland orrectness otthein pages parsin parsing patrick paul pereira pplication prague prec press proe ptvccssing pulturn purdom qcbraic raic rainer randall rase ratificationbased reasoning references related rend report resnan rimenta rmalisms rming rogers rollinger ronald sankt schemata schneider schrift science sclmeider seiflb sentation shiebcr sikkel springer structure stuart suny sysi system tableaux talking technologies technology text theories theory titor tran transition tree trees trent twentc twente twlt ulhmm understanding unification universiteit ustin vcessing viol walton warren wesley wilfi william with wmso woods work workshop york
http://acl.ldc.upenn.edu/C/C00/C00-1050.pdf	50	Kana-Kanji Conversion System with Input Support Based on Prediction	actors aguilera alice arbit automatically based built cagigas calculated candidates certainty character choice cjlavacter cmldid context corpora could dictionary dietio disappear each ends entered entry estimated experiments extracted factor field fill frequency from gnuson greater hess hunnicutt information input into japanese kinds lhtm likely naya next ones onversion operation operations original palazuelos phrases port positions precision prediction presented processor proposed raise rare rary reduce references registered rein released rlberger ropriate santiago shown shows sira special strillgs strings system texts than thatures this thresholds ticlionary timt tosiiiba typed used user using when which with without word words work zords
http://acl.ldc.upenn.edu/C/C00/C00-1080.pdf	80	Chart Parsing and Constraint Programming	abdemmdhcr abdennadheh about algorithms allow almlyse also analysis approach aspects associated attribute attributive automatic available based bday become benelils between both bottom carnegie carpenter chart chomsky classical collference collj combines combining compared compilation computational conclusion conference conlluence conqmter conqnttation constraint constraintli constraints context created ctjeacl current deduction deductive derivational descriptions different directly down draft drive each easy edge efficiently efliciency eflicient engine enough environment erence etacs example existing feature flmrefore form formedness framework frfihwirth friihwirth from generator gotten graham grammars handling harrison homogenous hpsg implement implemented implementing improved interleaving into invested itself iwpt journal lacking language languages larger leeuwen left linguistic linguistics llexible lnai lncs logic logical ludwig madrid mamtal manandhar matiasek maximiliansuniversitfit mellon meurers meuss mfinchen minimalism minimalist model modes morawiclz more natural nplemenlation offers operational operations other over paper parsers parsing particular patwing penn pereira phnciple phrase pittsburgh pmgtwmming pmgtzmlnting post posting powerful practice prag presented press principles proach problem proceedings process processing program programming propagation propose providing pure realization recognition recognizer references regelbasierten relational remedied resolution resulting retor right rther rules ruzzo schabes schema schemata science seem series sern shieber shown sikkel similarity simple smolka solvers solving some spain specialized specifically specification springer stabler strategies structures studies such suffer system systems tcstbed techniques teinach that thcory them theories these thesis they this tiibingen tions tlexibility typed types unification universal universitiit university used user using version warren whether which while wien wing with wnsac work workshop ying
http://acl.ldc.upenn.edu/C/C00/C00-2100.pdf	101	Automatic Extraction of Subcategorization Frames for Czech*	accurate acljsigdat acqttisition acquiring acquisition adapting adaptive aggregated akira alex alternations ambiguity annotated applications approaches artijical automatic autonaatic barbora basili bcategorization bergen berkeley berlin bickel boguraev brent brew briscoe building cambridge canada carroll case categories chamiak chemnitz chris christopher class classes classilication coincidence coling columbus computational comwctionist conference corpora corpus cues david dependency diathesis dictionary disambiguation distributions domain dunning ecml editors emnli empirical ersan eugene evans evidence extraction fagging features fiom flames frames frequencies frequency from furtg generalizations germany gibson gleinnan glenn grammar grammatical granada guido head help holden improvement induction information inllective instead intelligence issues istics itaji iven jacl john june kamin kariaeva karolinum kjell knowledge landau language languages lapata large lcxicalized learning learns lecture lexical lexicon ling lingmt lingui linguistics local lxmguage lztrge manning march marcus maria mathematical mats meaning meeting merlo methods michael michele minnen mitchell montrdal montreal morphological mort murat natalia natural nlpd norway notes ohio oksum omlmtational orpora pagcs pages paoht paola parser pascale pcfg peter prague praha prediction press probabilities proceedil proceedings processing program pustejovsky references reprinted resolve rich riloff roberto roceedings rocessing rooth scheler semantic sentence siglex spain springerverlag statistical statistics stevenson stic structured study subcategorisation subcategorization supervised surprise suzanne symbolic syntactic syntactically syntax systems tagged tagset tanlps temni text towards treebank ttaji ttladkfi universitd unsupervised untagged ushioda using valence valency verb verbs very vmdigni volume waibel washington wcrmter webster what whitehouse with workshop wvlc zhou
http://acl.ldc.upenn.edu/C/C00/C00-1005.pdf	5	An ontology of systematic relations for a shared grammar of Slavic	acci accommodating accord acquisition adjunction adjunkliotl agrccnlcnt agreement alism allgcmcinen alomizalion although ammal ananlar applications applied atlems avguslinova avgustinova bemovfi benefit between bilingt blake bolua borsley budapcsl bulgarian call calllo cambridge cambrktge cann case charts checking clilics cliticisation clitics closely clustering come comewashed comparative computational concord concordial consider considerhim constituents control corbcll corl cross csli dcscrip dependence devo dialhcsis differences disguiscd dojdeg durnia efficiency elllll expecl experience fdsl flickingcr fool fotu fraser from gained gamon girl grammar grammatical greatly gtjala have hcadedncss healthy historisehe hpsg iacc ianguislics ianlp icken iect iety ilagreementcaseinoml imnguagc indiana insights inst interference inullilingual iournal iqvss isscrlalions issue itpsg iukham ivan john kanleyama knowledge kongruenz lagrecmcnt language languages lavicae lbol lchfeldt lffficicnt lhcory linglish linguistic linguistics lllflrking lnguage lnst lozano lranslalion machine makes maria mark marking mary maskiraua matching mcglashan mction meeting miinchen milego mivan mmlyses mnch modelling models morphosynlax morphosyntax motiwltion nalural nego ngincering nicc nice nomi nominal ocpcn ofco okazalas oliwt opportunities order ordercd orderedhim original osition otto panevovfi paper parallelism parsing particle philologiae pinkham polish potsdam predication predicative prepositional press prijti problems proceedings processing psycholinguistic raclical rcli rclizrence rcnce reached redmond references rektion related relational rence rencing report reportedly research results reutler rkowski rocccdings roccedings roceedings russian russischen rzcpi saarbri sagner sample schmidl second secondary shared sharing similarilies simile sktlt slate slavic slovoso sltldy some special specimina stradala style subcatcgorisation subcategorisation such sufferedlbr sufl suited systcmalischc systems tcxlbooks technical technology thai that them these they tlcads treatment trod tsujii ttpsg turn turncd turned typological univcrsily universily university unlcrstlchtmgen useful uszkorci uszkoreit veleli verb vidjal volume washcd were will with word work workshol wortfiigungcn wouldcomedisguised york zdorovoj zrcnce
http://acl.ldc.upenn.edu/C/C00/C00-1075.pdf	75	Application of Analogical Modelling to Example Based Machine Translation	algorithm aligning analogical based boutsis clauses cognitive coling collins communications comparison computer conference correction cranias cunningham daelemans damerau daniel detection dublin durieux ebmt edited empirical errors example gillis harold international jones june language lazy learning machine matching methodology methods modelling natural papageorgiou parallel piperidis press proc processing references science skousen somers spelling technique texts translation with
http://acl.ldc.upenn.edu/C/C00/C00-2139.pdf	139	ABL: Alignment-Based Learning	aage abilist ability acoustical acquiring acquisition actic aistics algorithm almost although america andreas andrew annotated annual approach approaches apter artificial association asymptotically ation ational atis automata barker bascd based bayesian berlin best beyond bmmds bonnema booth building cambridge caraballo carl categories charniak chart chater chert chicago codes comlmtational communication compression compu compul computational computer computing comtmter conceptual connectionist context convoh corpora corpus correction cory csli cwnitivc daelemmls data debm decoding departnmnt describe dgorithm differeat distrilmtional ditl domain ebank editor editors electrical electronics engineering engineers english eory erent error ertn estimation eugene eurwpean fcrcnce figures finch first fischer flgori fllture formed free furthermore generalization generating gralmnar grammar grammars harris have ilistic iloif imes induction inference information ings inside instances institute intbrmation intelli intelligence journal kehler klatt lang language languages langv large larger lari learn learning lecture lengl lenth lexical lexicon lied ling linguistics lmlguage mach machinery madrid magerman marcinkiewicz marcken marcus martin massachusetts meetin memory merit methods michael model modeling mutual national natural near newark nfinimmn nick ninety notes oriinwald outside ovis pages papers parsing penn pereira peter plan pnbolic powerflfl preface press probal problem proceed proceedings processing prol promising publications pwcecdin qence quistics racketed rcnce recognition recursion redington reestimation references representation results robert rocccdings rocessing santorini scha scheler schgd science scription semantic sentences seventh sharon size small society sommerset speech springer ssoci stanjbrd stanley statistical statistics stefh steven stochastic stolcke stoleke string structural symposium synt tational technology theory these thesis tially timum tion tiona trainabh transactions translation university unsuper unsupervised used using verlag viscd viterbi vohmm vohnne wagner walter wernter witcl wolf wollf workshop young zellig zpcricncc
http://acl.ldc.upenn.edu/C/C00/C00-1038.pdf	38	Directional Constraint Evaluation in Optimality Theory*	adgc ainl aivation alan algorith aligmnent alized amherst andid ankara anwual apply arbitrarily archive armin aspects assimilation atmr auto bakovid barcelona beckman benjmnin beyond bilkent booij both boulder brmal bruce bruening cambridge change chicago clustell cmatik coling colmexion colorado comes coml comlmta comp compare complexity computal computational configured cons consl consla constr constraint constraints conventions cormen cory data dcsir decomposed description dirccl dirct directional disfavored ditors dobrin does douglas down duster each eaul editor edsger efficient eisner eiten ellbx empirically enclosing erativc eric evaluation even faithflflness figh finitc finite finsl fints flly fluation footlpoltm forln formal formul frank genca gener generalized generative giorgio glsa grat grid hcir hnve iaap iangv identity ijkstra iled ilnporl imality instan interaction into introduction iohn ional ionld ionnl ioual iously irectional isner istics itow jason jill jochen john johnson july kaplan karttunen kluwer language lciserson leaving level lfinda lfllonology lflmnological liml ling linguistic linguistics linguistik lmtterns local locations lombard lower madrid many mark markedness markus mart martin massachusetts math mations mccarthy mehryar mende mester metrical milil mints mlalysis models mohri mona morph morphology most mouton msduc note notion ntithfltlncss numcrisch offended offsmnlp ohvious olmhl ology ommed onstrainl optilmdity optimalit optimality oscd ounded papc papers parascssion phonological phonology positional potsdam power press primal primitive prince principles problems proc processing prol pros prosodic prot ptimality quebec raint ralizcd ralnnlar ranked rants rbook reduplicative references regular relating repres restrictions revisited riately rictly rimil rivest rlying rmlk robert rocedurality rolll rule rutgers scil scope shoukt showl singer smolensky society speech start state stemberger stnt stress studies style syllalfification systems tamlsducers tdth tesar theory these thesis this tinitc tion tiona tional tionin tirc titeory tlmory tone trammar treatment trommer ttayes turkey uage uasc unmarked utgers versions viii violability voicing walthe what where with work working worksh worse
http://acl.ldc.upenn.edu/C/C00/C00-2118.pdf	119	A u t o m a t i c Lexical A c q u i s i t i o n B a s e d on Statistical Distributions*	according acquiring acquisition agremcnt aims akira alld almer alternation alternations ambiguity amin ammatical anlp aola aolm argument athens attd aufmamt automatic behaviour bergen beth boiled boluiie bound branimir brew briscoe canada cation changcd chicago chris christopher class classes classification classifipredicate clca cleared cllinatsu collapsed college collins compatational complexity comput construction cooled corpora corpus cracked cues cxieal danced dang diana dictionary disambiguatlon dissolved distributions divided drops dtcd eallcd earvcd editors empirical encodillgs english enhage enmark ensions entries eport equency erence ersemantic essing etecting evin exica exploded extraction eyser features floated flooded folded foreign fractured frames frocs from galloped generative glided grammar greece hardcned hiked hopped hurhot idcntit igdat indicators information ingual inheritcd inter interlingual international intersective investigating itors james japanese jogged johm john joint jones journal judge jumped kariaeva karin kiekcd kipper knittcd langaage language lanuagc lapata large larlliall lation lcvin leaped learning lexical lexicallsed lexicon lhedieting linguisi linguistic linguistics ljuilding loseph machine madrid maehi manning mapping marched maria maryland mateo matsumoto mccarthy mcred melted merlo mett michac michael mirth models montreal morgan mort mouaffc multilingual natalia natural norway oased object oguraev ohio oishi onnie opcncd oraniscd organization osenzweig packcd pages paintcd paraded park participation playcd press procs pzilivc qlex quinlan quisition raced rcnted reaped references referenee regular relations rence representation resolve ried rocc rocessinq rocs rograms role roll rushcd sabine scale science scurricd seattle sehulte semanl semanrowed semantic sense sentence seootcd siegel simboltnie sing skclehe skipped sllzantle solidified spain stabilized stablishing state statistical stevehsoll stevenson structure studied stuttgart subcategorization subcategorlzation subclasses supervised suzalllle suzanne suzmme swallowed switchitlg syntactic syntax tale task texts their thematic thesis three tics tiptoed tisition toss trang translation turc tutoring typed uilding unaccusativcs undcrslandin unergatives universitst university unsul upper using ustetjovsky vancouver verb verbs view vised vttcd vveinberg walde wandered washcd washington webster whitehouse widened word ycllcd yuji
http://acl.ldc.upenn.edu/C/C00/C00-1082.pdf	82	Bunsetsu	able accent accuracy acem adam adwait aecm ages agichtein alld also although ambiguity anahdzer analogy analysis analyzed analyzer andrew anese answer application approach aring arklfi artificial artition artitioned artmcnt artment ased association assoeitation assumption atnaparkhi atnat avid based berger beta better betwee between borthwicl btmsetsu bunsetsu carried case cases category chapter chunks classitieation clauses clmnking combination companies computational comt conclusion convention corlms corpora corpus correct correctly cottference decision decisiontree della department dependency described diverse ecision ecoglfition eech ella elltropy empirical english entity entrol entropy eric erik eriments eugene exami example exclusive existing experiment exploiting extracting flile following from grishlnan grouping guage high highest hitelligence http identification iiitoshi iluman imnnetsu incit incorrectly increases indicates inear inform informatics information inserted inserting institute intorlll isalmra jalmnese japanese japmmse john journal kameda kazuhiko kiyotaka knowledge kurohashi kyoto language lapanese lapmmse lapnese lasts learning lebased lexieal lgurohashi linguistics list lists lligh llllsetsll lmarning lmme lorn lsllro machine makot makoto marcus mark masayuki matics matsumoto maxilnum maximmn maximum measur measure mechanical meeting metal method methods missed mitchell mlshaw mmther mnemonic model modeling models mpirieal nagao named national natural need ninth nishiokayama ntropy observed obtained oceedings oifly olcst only order other ozeki pages partitiol partition partitioned performed phemes phological pietra preference principles proceedings proceedira processing qrams quage quinl ralneworl ralph rate recall recision reenstra references release research resemvh resenting resolution resolving restoration revious ristad rivest roblem rocecdin rocessings roeessings ronald roof ropy rove rules sadao sang satoshi segmentation sekiim semantic sentences shigeyuld similarity simple sixth small software solve some sources space spanish statisticai stephen sterling structure subordinate sven system syuei syui tagging takehito tbur test text than that there this thod tics tile time tjong tllese tool toolkit transformationq transh tree trees ttion uase uehimoto ulfiversity ulwpeau university used uses using utsuro version very vincent were wgnl when which with word work workshop yarowsky yench yoto yujie zhang zlmng
http://acl.ldc.upenn.edu/C/C00/C00-2167.pdf	167	Japanese	across also among analyses analysis artifact artifacts based been believe beneficial best boundaries case committee comparable comparison conclusion contilme contribute could created definition degrades dependency described difference different difficulty ditficult ditficulty domain effort english etween evaluation ewdnation explained expression expressions fence finally find first fitcadm fllture future hasis have hitoshi hmnans holmflflly homepage hornepage html httl http hulnan human identifying improve include indexen introduced irex isahara jaist japan japanese kind level lmman lrec matured measure merged more nacsis newly next nist nnei ntcir numeric obvious opposed overall pattern perfbrlnance perform performance performed perfornlancc places poorly possibly proceedings project projects proteus reason reasons references reported result results satoshi sekine several side similar step sumlnarization superior system systems task technology than that there this tile time trec trove ttomei type types very well were where whereas which will with workshop
http://acl.ldc.upenn.edu/C/C00/C00-1063.pdf	63	Japanese Case Structure Analysis by U n s u p e r v i s e d C o n s t r u c t i o n of a Case Frame Dictionary	accuracy accurately acquisition agency akio alion allow alne also ambiguous analysis anese anlp annual around arsing association automatic based basic because briscoe building carroll case cases cave ceiling checked chnology christopher clausal close collstllllptioll compared components computational conj conjunctive considering consists construction corpora corpus correct correctly crcncc csources cxphdn detect detected detection dictionary differed easily editors eizai employs evaluation exmnt extraction fairly figure fimmclal foreign frame froln from general golden good guide hayashi head heuristic hifa hiromi icon iglppyolt ikehara illltl imln improvement improving incorrect increased information international iwananfi ixmshi japa japan japanese john kentarou kurohashi kyoto large learning lexicon linguistics long lvlltttl lwnguage markers masahiro matsumoto maturing mectin mexico mikka minislcr misaku miyata miyazaki mmlysis model modifier modifiers mulo nagao nakaiwa national obtained occcding ogura oredhy original overall pages parse parsed pllllcy policy preference prevention proceedings promotion proposed prvcccdin prvccedings publishing quistics references relations result results rlncc rule satoru satoshi scaled seems selection sentences shhula shirai shirl shown simeific since small snbcategorization standard strong structures subcategorization superior syntactic system table takashi takehito task tcsc that their then third this those though tile tmltli tnitonwru topic university unsupervised used using utsuro valuable verbs versa very vice were whether which while with wtite yokoo yoltha yoshiflmfi yoshihiko yuji
http://acl.ldc.upenn.edu/C/C00/C00-2089.pdf	90	Tagging and Chunking with Bigrams	abney above acadenfic accuracy acoustics alil almey also always ambiguity anal analyzer anlp annotated annual aper applied approach approaches aptcr argamon artificial assmnplion association atro august automatic automatically base based because berck berwick bourigault brill buchholz building cambridge canada cascades case changes chanod chemnitz chunking chunks chunldng church clarksond classifier clauses cmpora cncc coling colingacl colloquium combination combine comparison component computational conclusions confcr confcrcncc confcrcnce conference contextual copenhagen coping corpora cortms cutting czech daelelnans daelemans dagan data decision definitions denmark detector develol different difl direct disambiguation ditficult done dordrecht driven eacl ecml editor eech ejerhed eland elmblie emnlp empirical english entrol erent erfl eric error estimation european eurospccch evaluation extraction factor feasible finding finit finitary finite following from future generator germnly gging gillis grammars grammatical granada grmnmatical gtnmmatical halteren have higher hnproved homogeneous horacio hybrid icgi identification ieee igrams improving including incremental inference inodels integrated intelligence international involv joint june katz kinds kluwer knowledge krymolowski lafflc languagc language large learning learnt lecture lexical linguistics llufs lnai loth lower lrec lyansfbrmation machine made magerman marcinkiewicz marcus maryla maximum mccting meeting meinory memory mentioned merialdo meteer methods mhrquez mnong model modeling models mokhtar montrdal natural nddellee nnsactions noml notes noun nptool obviously ones ortable other padrd pages pahnueci parser parsi parsinfl parsing part partial parto parts patterns pederson penn pfvcccdings phrase phrases pierce pora possible practical preaches precision prieto principle probabilistic probabilities procccdinfls procccdings proceedings process processing program prunning prvcccdings prvecssing publishers quantitative rague ramshaw rate rates ratnapm rcscntcd rease reasons recall recognizer reece reference references resented resources restrict rhodes rmanc rminological roaches robust roceedi rodhguez ronsenfeld rouveirol santorini schwartz second section sequential series sets shallow signal similar similm sizes slightly some sources spain sparse speech springer state statistical stochastic stone structure study surface syntacti syntax system tagger tagging tags tbrmalism tcnny terms test text than that third this through tile tionary tlexibte tliciency tlmn toolkit train transibnnation treebank trees university unknown unrestricted used uses using vashington veenstra verlag very voutilainen washington weischedel with wordclass words work workshop zavrel
http://acl.ldc.upenn.edu/C/C00/C00-2119.pdf	120	Using a Broad-Coverage Parser for Word-Breaking in Japanese	above acquiring adcock alexis algorithm ambiguous amsterdam analysis analyzer andi annoyed antony appear appendix applications applied approach artificial assistance association attachment attgmented attice auxpi backward base based been benjamins best binary boston breaker breaking brockett building burstcin bvel carmen carroll cassandra chapter char charniak chmdia choice chris city classical coling commercial computational concept conji connective corpus correct cost country dale decli determining dissertation dolan down either eugene example experience figure first fitted forthcoming forward from fuchi gamon gary george glenn gotoh grammar grammars harold heidorn hektorn hermann hidden hisami hisamitsu history hnproving ieice incomplete inferring information input intelligence intelligent intervention interventions into ipsj issues japanese jensen jeremy jessie jiang jill john jtag kacmareik karen katz kluwer knowledge koichi kurohashi language lattice leacock learning lexical lfandbook linguistics literature littmaa local lozano lpsj lrec lucy making makoto manasler markov masaaki mathematics matsumoto mccann method michael microsoft mindnet minimum model modify moisl morphological multilingual murakami nagao nagata natural neighboring nitta noun nouna occurrence parameter parse parsed parsers parsing phrase pinkham plnlp posp pospi practical practice press proceeding proceedings processing proni publishing ramer references relatkms relcli report reseamh research results returned reutter richardson robert robust rules sadao sagayama schank search segmentation seiners semantic semantically sentence sharing shinichiro shown signlc similarity sponsored stephen stochastic structure structuring successive syntactically system taggers takagi takeshi taket technical text thai theoretical topic tradition tree trees university using vanderwende verb verbs webber while william with word work workshop writing york yoshihiko yuji zixin zuki
http://acl.ldc.upenn.edu/C/C00/C00-1018.pdf	18	The Use of Instrumentation	afro agnss alshawi analysis annotation annual balkan based beil btrean carroll carter ceder ciravegna cmnbridge computer conf corpus cruz dcsign developing eagles editors entropy esov essex estimation eval evaluation ewgpr fast final first flickinger fouvry generation gerlnan grammar granada grmmnar grmnmar inside international ixayner journal language langv lavelli lexicalized lization maryland mecting meetin meeting natural outside pages parsing petrelli pianesi plications prec prescher processing profiling rayner rces reesources references report resources riezler rooth runing sanmelsson santa scheme slli smnuelsson spain specia specialization speech spezialization spohen suitc suite systematic systems techn test thresholds through towards translator tsnlp uation university using with year years