资源描述:
《语义查询扩展中词语-概念相关度的计算》由会员上传分享,免费在线阅读,更多相关内容在行业资料-天天文库。
1、ISSN1000-9825,CODENRUXUEWE-mail:jos@iscas.ac.cnJournalofSoftware,Vol.19,No.8,August2008,pp.2043−2053http://www.jos.org.cnDOI:10.3724/SP.J.1001.2008.02043Tel/Fax:+86-10-62562563©2008byJournalofSoftware.Allrightsreserved.∗语义查询扩展中词语-概念相关度的计算1,2,31,2+1,2田萱,杜小勇,李海华1(教育部数据工程与知识工程重点实验室
2、,北京100872)2(中国人民大学信息学院,北京100872)3(北京林业大学信息学院,北京100083)ComputingTerm-ConceptAssociationinSemantic-BasedQueryExpansion1,2,31,2+1,2TIANXuan,DUXiao-Yong,LIHai-Hua1(KeyLaboratoryofDataEngineerandKnowledgeEngineerfortheMinistryofEducation,RenminUniversityofChina,Beijing100872,China)2(
3、SchoolofInformation,RenminUniversityofChina,Beijing100872,China)3(SchoolofInformationScience&Technology,BeijingForestryUniversity,Beijing100083,China)+Correspondingauthor:E-mail:duyong@ruc.edu.cnTianX,DuXY,LiHH.Computingterm-conceptassociationinsemantic-basedqueryexpansion.Journ
4、alofSoftware,2008,19(8):2043−2053.http://www.jos.org.cn/1000-9825/19/2043.htmAbstract:Insemantic-basedqueryexpansion,computingterm-conceptassociationisakeystepinfindingassociatedconceptstodescribetheneededquery.AmethodcalledK2CM(keywordtoconceptmethod)isproposedtocomputetheterm-
5、conceptassociation.InK2CM,theattachingrelationshipamongterm,documentandconcepttogetherwithterm-conceptco-occurrencerelationshipareintroducedtocomputeterm-conceptassociation.Theattachingrelationshipderivesfromthefactthatatermisattachedtosomeconceptsinannotatedcorpus,whereatermisi
6、nsomedocumentsandthedocumentsarelabeledwithsomeconcepts.Forterm-conceptco-occurrencerelationship,itisenhancedbythetextdistanceandthedistributionfeatureofterm-conceptpairincorpus.Experimentalresultsofsemantic-basedsearchonthreedifferentcorpusesshowthatcomparedwithclassicalmethods
7、,semantic-basedqueryexpansiononthebasisofK2CMcanimprovesearcheffectiveness.Keywords:semantic-basedqueryexpansion;concept;ontology;term-conceptassociation摘要:在基于语义的查询扩展中,为了找到描述查询需求语义的相关概念,词语-概念相关度的计算是语义查询扩展中的关键一步.针对词语-概念相关度的计算,提出一种K2CM(keywordtoconceptmethod)方法.K2CM方法从词语-文档-概念所属程度
8、和词语-概念共现程度两个方面来计算词语-概念相关度.词语-文档-概念所属程度来源于标注的文档集