词义消歧算法与应用_PDF下载[104MB-百度云]艾吉瑞

本书特色

[

《词义消歧—算法与应用(英文影印版）》是”计算语言学与语言科技原文丛书”中的一册。对于计算机来说，要理解人类语言就必须消除歧义，在计算语言学领域，词义消歧（word sense disambiguation，简称wsd）一直是研究者探索的内容。本书是近年来国际学术界关于词义消歧研究成果的一部集成之作。几乎覆盖了词义消歧研究各个题目，具有重要学术价值。

]

内容简介

[

《词义消歧—算法与应用(英文影印版）》是**本全面探讨词义消歧的书，对于重要的算法、方式、指标、结果、哲学问题和应用，都有涉略，并有这个领域的权威学者对本领域的历史及发展所做的较为全面的综述。研究者可以从本书了解到本领域的成果和发展趋势，开发人员可以从本书了解一些技术和方法。

]

导读contributorsforewordpreface1 introductioneneko agirre and philip edmonds1。1 word sense disambiguation1。2 a brief history of wsd research1。3 what is a word sense?1。4 applications of wsd1。5 basic approaches to wsd1。6 state-of-the-art performance1。7 promising directions1。8 overview of this bok1。9 further readingreferences 2 word sensesadam kilgarriff2。1 introduction2。2 lexicographers2。3 philosophy2。3。1 meaning is something you do2。3。2 the fregean tradition and reification2。3。3 two incompatible semantics?2。3。4 implications for word senses2。4 lexicalization2。5 corpus evidence2。5。1 lexicon size2。5。2 quotations2。6 conclusion2。7 further readingacknowledgmentsreferences 3 making sense about sensenancy ide and yorick wilks3。1 introduction3。2 wsd and the lexicographers3。3 wsd and sense inventories3。4 nlp applications and wsd3。5 what level of sense distinctions do we need for nlp, if any?3。6 what now for wsd?3。7 conclusionreferences 4 evaluation of wsd systemsmartha palmer, hwee tou ng and hoa trang dang4。1 introduction4。1。1 terminology4。1。2 overview4。2 background4。2。1 wordnet and semcor4。2。2 the line and interest corpora4。2。3 the dso corpus4。2。4 open mind word expert4。3 evaluation using pseudo-words4。4 senseval evaluation exercises4。4。1 senseval-1evaluation and scoring4。4。2 senseval-2english all-words taskenglish lexical sample task4。4。3 comparison of tagging exercises4。5 sources of inter-annotator disagreement4。6 granularity of sense: groupings for wordnet4。6。1 criteria for wordnet sense grouping4。6。2 analysis of sense grouping4。7 senseval-34。8 discussionreferences 5 knowledge-based methods for wsdrada mihalcea5。1 introduction5。2 lesk algorithm5。2。1 variations of the lesk algorithmsimulated annealingsimplified lesk algorithmaugmented semantic spacessummary5。3 semantic similarity5。3。1 measures of semantic similarity5。3。2 using semantic similarity within a local context5。3。3 using semantic similarity within a global context5。4 selectional preferences5。4。1 preliminaries: learning word-to-word relations5。4。2 learning selectional preferences5。4。3 using selectional preferences5。5 heuristics for word sense disambiguation5。5。1 most frequent sense5。5。2 one sense per discourse5。5。3 one sense per collocation5。6 knowledge-based methods at senseval-25。7 conclusionsreferences 6 unsupervised corpus-based methods for wsdted pedersen6。1 introduction6。1。1 scope6。1。2 motivationdistributional methodstranslational equivalence6。1。3 approaches6。2 type-based discrimination6。2。1 representation of context6。2。2 algorithmslatent semantic analysis （lsa）hyperspace analogue to language （hal）clustering by committee （cbc）6。2。3 discussion6。3 token-based discrimination6。3。1 representation of context6。3。2 algorithmscontext group discriminationmcquitty’s similarity analysis6。3。3 discussion6。4 translational equivalence6。4。1 representation of context6。4。2 algorithms6。4。3 discussion6。5 conclusions and the way forwardacknowledgmentsreferences 7 supervised corpus-based methods for wsd8 knowledge sources for wsd9 automatic acquisition of lexical information and examples10 domain-specific wsd11 wsd in nlp applications