{"id":"https://openalex.org/W2767340342","doi":null,"title":"Recognizing textual entailment using deep learning techniques","display_name":"Recognizing textual entailment using deep learning techniques","publication_year":2017,"publication_date":"2017-07-06","ids":{"openalex":"https://openalex.org/W2767340342","mag":"2767340342"},"language":"en","primary_location":{"id":"pmh:oai:dnet:upcommonspor::1d81ee512b61368e385e43d2d98fd795","doi":null,"is_oa":true,"landing_page_url":"http://hdl.handle.net/2117/109802","pdf_url":"http://hdl.handle.net/2117/109802","source":{"id":"https://openalex.org/S4306402641","display_name":"LA Referencia (Red Federada de Repositorios Institucionales de Publicaciones Cient\u00edficas)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I4383465926","host_organization_name":"LA Referencia","host_organization_lineage":["https://openalex.org/I4383465926"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"info:eu-repo/semantics/masterThesis"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"http://hdl.handle.net/2117/109802","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5082137392","display_name":"Yang Han","orcid":"https://orcid.org/0000-0003-3782-4655","observed_orcids":["https://orcid.org/0000-0003-3782-4655"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Yang Han","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0003-3782-4655","affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5082137392"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9096999764442444,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9096999764442444,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.0674000009894371,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.005400000140070915,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/natural-language-inference","display_name":"natural language inference","score":0.9990000128746033},{"id":"https://openalex.org/keywords/sentence-pair-classification","display_name":"sentence pair classification","score":0.4350000023841858},{"id":"https://openalex.org/keywords/character-level-embeddings","display_name":"character embeddings","score":0.40700000524520874},{"id":"https://openalex.org/keywords/attention-mechanism","display_name":"attention mechanism","score":0.335999995470047},{"id":"https://openalex.org/keywords/chinese-named-entity-recognition","display_name":"Chinese named entity recognition","score":0.47099998593330383},{"id":"https://openalex.org/keywords/chinese-sentiment-analysis","display_name":"Chinese sentiment analysis","score":0.5189999938011169},{"id":"https://openalex.org/keywords/chinese-text-summarization","display_name":"Chinese text summarization","score":0.4009999930858612}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.755906343460083},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.7390012145042419},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6980776786804199},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.6798063516616821},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6669352054595947},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.5831655859947205},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5754510164260864},{"id":"https://openalex.org/C95318506","wikidata":"https://www.wikidata.org/wiki/Q6588467","display_name":"Textual entailment","level":3,"score":0.5015885829925537},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4574545919895172},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.44791847467422485},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.44005846977233887},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3960316777229309},{"id":"https://openalex.org/C134752490","wikidata":"https://www.wikidata.org/wiki/Q374182","display_name":"Logical consequence","level":2,"score":0.3750874996185303},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3257898986339569},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.15765288472175598},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"pmh:oai:dnet:upcommonspor::1d81ee512b61368e385e43d2d98fd795","doi":null,"is_oa":true,"landing_page_url":"http://hdl.handle.net/2117/109802","pdf_url":"http://hdl.handle.net/2117/109802","source":{"id":"https://openalex.org/S4306402641","display_name":"LA Referencia (Red Federada de Repositorios Institucionales de Publicaciones Cient\u00edficas)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I4383465926","host_organization_name":"LA Referencia","host_organization_lineage":["https://openalex.org/I4383465926"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"info:eu-repo/semantics/masterThesis"},{"id":"pmh:oai:recercat.cat:2072/310155","doi":null,"is_oa":true,"landing_page_url":"http://hdl.handle.net/2117/109802","pdf_url":null,"source":{"id":"https://openalex.org/S4306402147","display_name":"RECERCAT (Consorci de Serveis Universitaris de Catalunya)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I4210090028","host_organization_name":"Consorci de Serveis Universitaris de Catalunya","host_organization_lineage":["https://openalex.org/I4210090028"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"info:eu-repo/semantics/masterThesis"},{"id":"pmh:oai:upcommons.upc.edu:2117/109802","doi":null,"is_oa":true,"landing_page_url":"http://hdl.handle.net/2117/109802","pdf_url":null,"source":{"id":"https://openalex.org/S4377196262","display_name":"UPCommons institutional repository (Universitat Polit\u00e8cnica de Catalunya)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I9617848","host_organization_name":"Universitat Polit\u00e8cnica de Catalunya","host_organization_lineage":["https://openalex.org/I9617848"],"host_organization_lineage_names":[],"type":"repository"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Master thesis"},{"id":"mag:2767340342","doi":null,"is_oa":false,"landing_page_url":"https://upcommons.upc.edu/handle/2117/109802","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":null}],"best_oa_location":{"id":"pmh:oai:dnet:upcommonspor::1d81ee512b61368e385e43d2d98fd795","doi":null,"is_oa":true,"landing_page_url":"http://hdl.handle.net/2117/109802","pdf_url":"http://hdl.handle.net/2117/109802","source":{"id":"https://openalex.org/S4306402641","display_name":"LA Referencia (Red Federada de Repositorios Institucionales de Publicaciones Cient\u00edficas)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I4383465926","host_organization_name":"LA Referencia","host_organization_lineage":["https://openalex.org/I4383465926"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"info:eu-repo/semantics/masterThesis"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2767340342.pdf","grobid_xml":"https://content.openalex.org/works/W2767340342.grobid-xml"},"referenced_works_count":19,"referenced_works":["https://openalex.org/W1843891098","https://openalex.org/W1924770834","https://openalex.org/W2064675550","https://openalex.org/W2095705004","https://openalex.org/W2131494463","https://openalex.org/W2133564696","https://openalex.org/W2136848157","https://openalex.org/W2163605009","https://openalex.org/W2170973209","https://openalex.org/W2267186426","https://openalex.org/W2292633562","https://openalex.org/W2470673105","https://openalex.org/W2607892599","https://openalex.org/W2951008357","https://openalex.org/W2952230511","https://openalex.org/W2953084091","https://openalex.org/W2963069010","https://openalex.org/W2964121744","https://openalex.org/W2964308564"],"related_works":["https://openalex.org/W2561577985","https://openalex.org/W7008650972","https://openalex.org/W2738049737","https://openalex.org/W3081241900","https://openalex.org/W2921109705","https://openalex.org/W2736621900","https://openalex.org/W2814865076","https://openalex.org/W2603048579","https://openalex.org/W2118463056","https://openalex.org/W7162725444"],"abstract_inverted_index":{"Textual":[0],"Entailment":[1],"(TE)":[2],"or":[3,43],"Natural":[4,601],"Language":[5,602],"Inference":[6],"(NLI)":[7],"refers":[8],"to":[9,41,263,276,334,375,382,418,457,491,536,630,654,685],"the":[10,28,45,61,115,173,181,212,222,243,259,265,272,278,290,301,322,332,343,349,365,370,390,398,406,416,438,455,462,465,483,498,509,512,516,524,527,541,562,570,583,588,604,615,628,643,649,657],"problem":[11,55],"of":[12,64,118,149,282,300,345,408,427,430,451,464,502,511,519,526,611,648],"determining":[13],"a\\ndirectional":[14],"relation":[15,46,544],"between":[16,47,545],"two":[17,546],"text":[18],"fragments.":[19],"To":[20],"specify,":[21],"given":[22],"a":[23,48,53,82,119,234,239,324,339,488,557,592,623],"sentence":[24,83,325,633],"pair":[25],"(a,":[26],"b),":[27],"task":[29],"is":[30,34,39,52,67,231,286,373,434,454,487,505,550,622],"to\\npredict":[31],"whether":[32,44],"b":[33,38,50],"entailed":[35],"by":[36,175,215,294,514],"a,":[37,42],"contradicted":[40],"and":[49,78,134,137,198,238,246,269,422,477,554,639,695],"is\\nneutral.\\nNLI":[51],"central":[54],"in":[56,94,152,348,364,394,530,619,634],"natural":[57],"language":[58,691],"understanding.":[59],"Recently,":[60],"dominating":[62],"trend":[63],"works":[65],"for\\nNLI":[66],"based":[68,104,169],"on":[69,187,200,587],"artificial":[70,88],"neural":[71,89,102,163,665],"networks,":[72],"which":[73,171,330,401,533],"aims":[74],"at":[75],"building":[76],"deep":[77],"complex":[79],"encoder":[80,625,687],"to\\ntransform":[81],"into":[84],"encoded":[85,521],"vectors.":[86],"End-to-end":[87],"networks":[90],"have":[91,146,184],"reached":[92],"state-ofthe-art\\nperformance":[93],"NLI":[95,226],"field.":[96],"For":[97],"instance,":[98],"there":[99,352],"are":[100,128,161,326,361],"recurrent":[101],"network":[103,164,262,393,666],"encoders,":[105,170],"which\\nrecursively":[106],"concatenate":[107,172],"each":[108],"word":[109,310,367,678],"with":[110,233,308,578,582,594,636],"its":[111,659],"previous":[112],"memory,":[113],"until":[114],"whole":[116],"information":[117,377,476],"sentence\\nhas":[120],"been":[121],"derived.":[122],"The":[123,228,448,566],"most":[124],"common":[125],"RNN":[126],"encoders":[127,303,333],"Long":[129],"Short-Term":[130],"Memory":[131],"Networks":[132,599],"(LSTM;\\nHochreiter":[133],"Schmidhuber,":[135],"1997)":[136],"Gated":[138],"Recurrent":[139],"Unit":[140],"(Cho":[141],"et":[142,157,166,191,205,218,296,316,445],"al.,":[143,158,167,192,206,219,297,317,446],"2014).":[144,318],"RNNs":[145],"surpassed\\nthe":[147],"performance":[148],"traditional":[150],"baselines":[151],"many":[153],"NLP":[154,202],"tasks":[155,203],"(Dai":[156],"2015).":[159],"There":[160],"also\\nconvolutional":[162],"(LeCun":[165],"1989)":[168],"sentence\\ninformation":[174],"applying":[176],"multiple":[177],"convolving":[178],"filters":[179],"over":[180],"sentence.":[182,340],"CNNs":[183],"achieved":[185],"state-ofthe-art\\nresults":[186],"computer":[188],"vision":[189],"(Krizhevsky":[190],"2012),":[193],"machine":[194],"translation":[195],"(Costa-juss\u00e0":[196],"M.R.,\\n2016)":[197],"also":[199],"various":[201],"(Collobert":[204],"2011).\\nIn":[207],"this":[208,493,612,620],"paper,":[209],"we":[210,257,270,288,436,507],"use":[211,258,271,304,389,472,482],"model":[213,224,230,245,285,396,414,433,513,539,572,617,650],"introduced":[214],"(Adina":[216],"Williams":[217],"2017)":[220],"as":[221,313,342,608,690],"baseline":[223,229,244,584],"for\\nthe":[225],"task.":[227,660],"consisted":[232],"word-level":[235,305,371],"embedding":[236,267,306,368,372],"layer":[237,275],"BiLSTM\\nencoder.":[240],"We":[241,388],"augment":[242],"propose":[247],"our":[248,254,283,395,413,431],"Character-level":[249,596],"Intra":[250,597],"Attention":[251,486,598],"Networks\\n(CIAN).":[252],"In":[253,319],"CIAN":[255,284,432,549,571,616],"model,":[256],"character-level":[260,291,391,399,635],"convolutional":[261,292,392],"replace":[264],"standard\\nword-level":[266],"layer,":[268],"intra":[273,439],"attention":[274,440,452,473,503,517,528],"capture":[277],"intra-sentence":[279],"semantics.\\nOne":[280],"contribution":[281,429],"that":[287,320,360,435,469,506,626,651,667],"implement":[289,437],"network\\nintroduced":[293],"(Kim":[295],"2016).":[298],"Most":[299],"sequence":[302,624],"layer\\ninitialized":[307],"pre-trained":[309,366,677],"vectors":[311],"such":[312,689],"GloVe":[314],"(Pennington":[315],"way,":[321],"words\\nin":[323],"not":[327,362,471,669],"independent":[328],"anymore,":[329],"helps":[331,534],"catch":[335],"more":[336,355],"internal":[337],"information\\nof":[338],"However,":[341],"growth":[344],"vocabulary":[346],"size":[347,463],"modern":[350],"corpus,":[351],"will":[353,402,474],"be":[354,403,682],"and\\nmore":[356],"out-of-vocabulary":[357],"(OOV)":[358],"words":[359],"presented":[363,618],"vector.\\nAs":[369],"blind":[374],"subword":[376],"(e.g.":[378],"morphemes),":[379],"it":[380],"leads":[381],"high\\nperplexities":[383],"for":[384,569,600],"those":[385],"OOV":[386],"words.":[387],"to\\nexploit":[397],"information,":[400],"computed":[404],"from":[405,425],"characters":[407],"corresponding\\nword.":[409],"By":[410],"doing":[411],"so,":[412],"gains":[415],"ability":[417,456,629],"learn":[419],"rich":[420,637],"semantic":[421,638],"orthographic":[423,640],"features":[424],"the\\nencoding":[426],"characters.\\nAnother":[428],"mechanism":[441,453,504],"introduced\\nby":[442],"(Z.":[443],"Yang":[444],"2017).":[447],"major":[449],"advantage":[450,501],"efficiently":[458],"encode\\nlong":[459],"sentences.":[460],"As":[461,661],"input":[466],"grows,":[467],"models":[468],"do":[470],"miss":[475],"precision":[478],"if":[479],"they":[480],"only":[481],"final":[484],"representation.":[485],"clever":[489],"way":[490],"fix":[492],"issue":[494],"and\\nexperiments":[495],"indeed":[496],"confirm":[497],"intuition.":[499],"Another":[500],"can\\nenhance":[508],"interpretability":[510,647],"visualizing":[515],"weights":[518,529],"an":[520,609,663],"sentence.\\nWe":[522],"conduct":[523],"visualization":[525],"chapter":[531],"5,":[532],"us":[535],"understand":[537,655],"how\\nthe":[538],"judges":[540],"textual":[542],"entailment":[543],"sentences.\\nThe":[547],"proposed":[548],"implemented":[551],"using":[552],"Keras":[553],"evaluated":[555],"upon":[556,573],"newly":[558],"published":[559],"MNLI":[560],"corpus\\nin":[561],"RepEval":[563,605],"2017":[564,606],"workshop.":[565],"test":[567,575],"accuracy":[568],"matched":[574],"dataset":[576],"is\\nimproved":[577],"0.9":[579],"percent":[580],"compared":[581],"model.":[585],"Based":[586],"improved":[589],"result,":[590],"we\\npublished":[591],"paper":[593,621],"title":[595],"Inference\\nin":[603],"workshop,":[607],"achievement":[610],"thesis.\\nTo":[613],"summarize,":[614],"has":[627],"to\\nencode":[631],"long":[632],"features.":[641],"Also,":[642],"attention\\nmechanism":[644],"provides":[645],"high":[646],"allows":[652],"people":[653],"how":[656],"model\\ndoing":[658],"it\u2019s":[662],"end-to-end":[664],"does":[668],"need":[670],"any":[671],"specific":[672],"pre-processing":[673],"or\\noutside":[674],"data":[675],"like":[676],"embeddings.":[679],"It":[680],"can":[681],"easily":[683],"applied":[684],"other":[686],"architecture\\ntasks":[688],"modeling,":[692],"sentiment":[693],"analysis":[694],"question":[696],"answering.":[697]},"counts_by_year":[],"updated_date":"2026-11-11T12:16:26.678958","created_date":"2017-11-17T00:00:00"}