An Introduction to Information Retrieve

《An Introduction to Information Retrieve》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、AnIntroductiontoInformationRetrievalDraftofNovember17,2007Preliminarydraft(c)2007CambridgeUPPreliminarydraft(c)2007CambridgeUPAnIntroductiontoInformationRetrievalChristopherD.ManningPrabhakarRaghavanHinrichSchützeCambridgeUniversityPressCambridge,EnglandPreliminar

2、ydraft(c)2007CambridgeUPDRAFT!DONOTDISTRIBUTEWITHOUTPRIORPERMISSION©2007CambridgeUniversityPressByChristopherD.Manning,PrabhakarRaghavan&HinrichSchützePrintedonNovember17,2007Website:http://www.informationretrieval.org/Comments,corrections,andotherfeedbackmostwelc

3、omeat:informationretrieval@yahoogroups.comPreliminarydraft(c)2007CambridgeUPDRAFT!©November17,2007CambridgeUniversityPress.Feedbackwelcome.vBriefContents1Booleanretrieval12Thetermvocabularyandpostingslists193Dictionariesandtolerantretrieval494Indexconstruction675I

4、ndexcompression856Scoring,termweightingandthevectorspacemodel1097Computingscoresinacompletesearchsystem1358Evaluationininformationretrieval1519Relevancefeedbackandqueryexpansion17710XMLretrieval19511Probabilisticinformationretrieval21912Languagemodelsforinformatio

5、nretrieval23713TextclassificationandNaiveBayes25314Vectorspaceclassification28915Supportvectormachinesandmachinelearningondocuments31916Flatclustering34917Hierarchicalclustering37718Matrixdecompositionsandlatentsemanticindexing40319Websearchbasics42120Webcrawlingand

6、indexes44321Linkanalysis461Preliminarydraft(c)2007CambridgeUPPreliminarydraft(c)2007CambridgeUPDRAFT!©November17,2007CambridgeUniversityPress.Feedbackwelcome.viiContentsListofTablesxvListofFiguresxviiTableofNotationxxvPrefacexxix1Booleanretrieval11.1Anexampleinfor

7、mationretrievalproblem31.2Afirsttakeatbuildinganinvertedindex61.3ProcessingBooleanqueries101.4TheextendedBooleanmodelversusrankedretrieval141.5Referencesandfurtherreading172Thetermvocabularyandpostingslists192.1Documentdelineationandcharactersequencedecoding192.1.1

8、Obtainingthecharactersequenceinadocument192.1.2Choosingadocumentunit202.2Determiningthevocabularyofterms222.2.1Tokenization222.2.2Droppingcommonterms:st