剑桥Information retrieval(信息检索)课件2.pdf

剑桥Information retrieval(信息检索)课件2.pdf

ID:52286766

大小:366.38 KB

页数:29页

时间:2020-03-26

剑桥Information retrieval(信息检索)课件2.pdf_第1页
剑桥Information retrieval(信息检索)课件2.pdf_第2页
剑桥Information retrieval(信息检索)课件2.pdf_第3页
剑桥Information retrieval(信息检索)课件2.pdf_第4页
剑桥Information retrieval(信息检索)课件2.pdf_第5页
资源描述:

《剑桥Information retrieval(信息检索)课件2.pdf》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、DRAFT!©April1,2009CambridgeUniversityPress.Feedbackwelcome.192ThetermvocabularyandpostingslistsRecallthemajorstepsininvertedindexconstruction:1.Collectthedocumentstobeindexed.2.Tokenizethetext.3.Dolinguisticpreprocessingoftokens.4.Indexthedocumentsthateachtermoccursin.Inthischap

2、terwefirstbrieflymentionhowthebasicunitofadocumentcanbedefinedandhowthecharactersequencethatitcomprisesisdetermined(Section2.1).Wethenexamineindetailsomeofthesubstantivelinguis-ticissuesoftokenizationandlinguisticpreprocessing,whichdeterminethevocabularyoftermswhichasystemuses(Sect

3、ion2.2).Tokenizationistheprocessofchoppingcharacterstreamsintotokens,whilelinguisticprepro-cessingthendealswithbuildingequivalenceclassesoftokenswhicharethesetoftermsthatareindexed.IndexingitselfiscoveredinChapters1and4.Thenwereturntotheimplementationofpostingslists.InSection2.3

4、,weexamineanextendedpostingslistdatastructurethatsupportsfasterquery-ing,whileSection2.4coversbuildingpostingsdatastructuressuitableforhandlingphraseandproximityqueries,ofthesortthatcommonlyappearinbothextendedBooleanmodelsandontheweb.2.1Documentdelineationandcharactersequencede

5、coding2.1.1ObtainingthecharactersequenceinadocumentDigitaldocumentsthataretheinputtoanindexingprocessaretypicallybytesinafileoronawebserver.Thefirststepofprocessingistoconvertthisbytesequenceintoalinearsequenceofcharacters.ForthecaseofplainEn-glishtextinASCIIencoding,thisistrivial

6、.ButoftenthingsgetmuchmoreOnlineedition(c)2009CambridgeUP202Thetermvocabularyandpostingslistscomplex.Thesequenceofcharactersmaybeencodedbyoneofvarioussin-glebyteormultibyteencodingschemes,suchasUnicodeUTF-8,orvariousnationalorvendor-specificstandards.Weneedtodeterminethecorrecten

7、-coding.Thiscanberegardedasamachinelearningclassificationproblem,asdiscussedinChapter13,1butisoftenhandledbyheuristicmethods,userselection,orbyusingprovideddocumentmetadata.Oncetheencodingisdetermined,wedecodethebytesequencetoacharactersequence.Wemightsavethechoiceofencodingbecau

8、seitgivessomeevidenceaboutwhatlan-guagethedocum

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。