Toward Large-Scale Information Retrieval Using Latent Semant(11)

2020-12-24 20:52

I am deeply indebted to Dr. Michael Berry, my major advisor, for his kind guidance and support. I also thank Dr. Susan Dumais, director of the Information Sciences Research Group at Bellcore, for her technical advice. In addition, she graciously allowed us

Traditionallexical(orBoolean)retrievaltechniques,whilesometimesvaluabletoex-pertstrainedtosearchcollectionsfromaspeci cdiscipline,oftenreturntoomuchinformationtotheuser.Othertimes,becausethetermsusedinthequerydifferfromthetermsusedinthedocument,valuableinformationisneverfoundinthedocumentcollection.

LatentSemanticIndexing(LSI)[DDF90],avector-spaceapproachtoconceptualinformationretrieval,isusefulinsituationswheretraditionallexicalinformationre-trievalapproachesfail.LSIestimatesthesemanticcontentofthedocumentsinacollectionandusesthatestimatetorankthedocumentsinorderofdecreasingrele-vancetoauser’squery.Sincethesearchisbasedontheconceptscontainedinthedocumentsratherthanthedocument’sconstituentterms,LSIcanretrievedocumentsrelatedtoauser’squeryevenwhenthequeryandthedocumentsdonotshareanycommonterms.Also,sinceLSIranksthedocumentsaccordingtotheirrelevancetotheuser’squery,thesystemhelpstheuserdecidewhichinformationmaybemorespeci ctotheuser’sinterests.

AlthoughLSIiscapableofachievingsigni cantretrievalperformancegainsoverstandardlexicalretrievaltechniques(see[Dum91]),thecomplexityoftheLSImodeloftencausesitsexecutionef ciencytolagfarbehindtheexecutionef ciencyofthesimpler,Booleanmodels,especiallyonlargedatasets.BycarefullyexaminingtheLSImodelandnotingthevariousoptimizationsthatcanbeappliedtoitsunderlyingimplementation,though,boththeretrievalbene tsoftheLSImodelandanexecu-tionef ciencynearthatoftheBooleanretrievaltechniquescanbeattained.Here,anef cient,extensible,maintainable,andportableimplementationoftheLSImodelispresented,andasimpleuserinterface,createdwiththenewimplementationoftheLSImodel,http://www.77cn.com.cningboththenewimplementationoftheLSImodelanditscorrespondinguserinterface,userscanquicklysearchlargedatasetswithoutunderstandinganydetailsoftheLSImodelorimplementation.

2


Toward Large-Scale Information Retrieval Using Latent Semant(11).doc 将本文的Word文档下载到电脑 下载失败或者文档不完整,请联系客服人员解决!

下一篇:垃圾收集路线设计

相关阅读
本类排行
× 注册会员免费下载(下载后可以自由复制和排版)

马上注册会员

注:下载文档有可能“只有目录或者内容不全”等情况,请下载之前注意辨别,如果您已付费且无法下载或内容有问题,请联系我们协助你处理。
微信: QQ: