基於组合特徵的汉语名词词义消歧

基於组合特徵的汉语名词词义消歧

ID:9292909

大小:177.50 KB

页数:12页

时间:2018-04-26

基於组合特徵的汉语名词词义消歧_第1页
基於组合特徵的汉语名词词义消歧_第2页
基於组合特徵的汉语名词词义消歧_第3页
基於组合特徵的汉语名词词义消歧_第4页
基於组合特徵的汉语名词词义消歧_第5页
资源描述:

《基於组合特徵的汉语名词词义消歧》由会员上传分享,免费在线阅读,更多相关内容在教育资源-天天文库

1、基於組合特徵的漢語名詞詞義消歧11基於組合特徵的漢語名詞詞義消歧本研究得到中國973重點基礎研究項目“面向新聞領域的漢英機器翻譯系統”(G1998030507-4)的支持。AStudyonNounSenseDisambiguationBasedonSyntagmaticFeatures王惠WANGHuiEmail:chswh@nus.edui.sgAbstractWordsensedisambiguation(WSD)playsanimportantroleinmanyareasofnaturallanguageproc

2、essing,suchasmachinetranslation,informationretrieval,sentenceanalysis,andspeechrecognition.ResearchonWSDhasgreattheoreticalandpracticalsignificance.ThemainpurposesofthisstudyweretostudythekindofknowledgethatisusefulforWSD,andtoestablishanewWSDmodelbasedonsyntagma

3、ticfeatures,whichcanbeusedtodisambiguatenounsenseinMandarinChineseeffectively.Closecorrelationhasbeenfoundbetweenlexicalmeaninganditsdistribution.Accordingtoastudyinthefieldofcognitivescience[Choueka,1983],peopleoftendisambiguatewordsenseusingonlyafewotherwordsin

4、agivencontext(frequentlyonlyoneadditionalword).Thus,therelationshipsbetweenonewordandotherscanbeeffectivelyusedtoresolveambiguity.Basedonadescriptivestudyofmorethan4,000Chinesenounsenses,amulti-levelframeworkofsyntagmaticanalysiswasdesignedtodescribethesyntactica

5、ndsemanticconstraintsofChinesenouns.Allofthesepolysemenounsweresurveyed,anditwasfoundthatdifferentsenseshavedifferentandcomplementarydistributionsatthesyntaxand/orcollocationlevels.基於組合特徵的漢語名詞詞義消歧11ThisservedasafoundationforestablishinganWSDmodelbyusinggrammatica

6、linformationandathesaurusprovidedbylinguists.ThemodelusestheGrammaticalKnowledge-baseofContemporaryChinese[YuShiwenetal.2002]asoneofitsmainmachine-readabledictionaries(MRDs).ItcanproviderichgrammaticalinformationfordisambiguationofChineselexicons,suchasparts-of-s

7、peech(POS)andsyntaxfunctions.AnotherresourceofthemodelistheSemanticDictionaryofContemporaryChinese[WangHuietal.1998],whichprovidesathesaurusandsemanticcollocationinformationofmorethan20,000nouns.Theywereemployedtoanalyze635Chinesepolysemousnouns.Bymakingfulluseof

8、thesetwoMRDresourcesandaverylargePOS-taggedcorpusofMandarinChinese,amulti-levelWSDmodelbasedonsyntagmaticfeatureswasdeveloped.Theexperimentdescribedattheendoft

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。