半格式化網(wǎng)頁信息提取與應(yīng)用.pdf

1、中國科學(xué)技術(shù)大學(xué)碩士學(xué)位論文半格式化網(wǎng)頁信息提取與應(yīng)用姓名：于魯波申請學(xué)位級別：碩士專業(yè)：信號與信息處理指導(dǎo)教師：俞能海邵正榮20080501Abs仃actABSTRACTWiththerapiddevelopmentoftheIntemetfasta11daccurateaccesstoinfomlationbecomesthebonleneckofthegrowingofindustryInte：met嬲the、^，orld’sla

2、rgeSt仃easureofinfomationresourceanractsmo佗跚dmorea牡entionThus，generalsearchengineappearsHowever，becauseofmeinfo咖ationoverJoadedproblem，itwillbringuselessinfo眥ationt0userS，verticalSearchhasbecomeanewhotspotBeingdi丘erentfo橢

3、generalsearch，verticaIsearchenginesonlyfocusoncertaina他asThrou曲initialextractionofwebinf0rmatioIl’ontheonehand，itcanimprovetheaccuracyofsearch；ontheomerhandbrowserSdonothaVeteadpagebypage，improVingreamnge仃iciencyThemainc

4、ontentOfthisthesisistechnologiesofconstructionontheverticaIsearchInmisthesis，welayemphasis0ndiscussiontheisSuesoftheEngIishwebpagekeywordex觚ctjon觚dproductwebpageinf0咖ationex燈actionThemainworkandinnovationsofthisthesisare

5、asfollows：1ProposeanEnglish、Ⅳebke”舊rdseXtractionalgorithmwhichisb硒edonpartofspeechg啪ularityThemethodfirStlyuses3一gramt0segmentsentenceint0phrases，gettingridofstopwordsatthes鋤etime；thengivesdifrerentmarkt0thecenterwordsan

6、dmordified、Ⅳords；atlastexportske”ⅣordsmroughsoftthresholdtechnologySowegenerate鋤uniVerSalkeyWordsextr：actionsystembasedonp礬ofspeechgranulari夠2ProposeanewSegmentationalgorithmf0fcommoditywebpageinfo咖atbnblocl(SThemethodis

7、basedon、ⅣebpageregularitystandardsproposedbythisthesisThethesisalsogivesthemathematicalexpressionofwebregularitystandardsInmisthesis，wefirstlyuseDOMtreet0representawebpage，XR氣：IH鼬mcturet0denoteaIeafnode；secondlycomputeth

8、ewebregular時；atlastthroughdoingtleXP!fqlHpathclusteringandthegettedwebregularityStatistic，thealgorithmcombinesheuristicrules，segmentSproductjnfomatjonbJocJ(S3ImplementaproductinfomationseXtractionalgorithmwhichiscombinat

9、ionofSegmentationmethodandwrappertooIThemethodappliedinthethesisfirStlyuseswebinfornlationblocksSegmentationalgorithmtosegmentproductinfo姍ationbIocl(S；thenutiIizesinstancecoVerageaIgorimmwhichisbasedontheDOMtoIe鋤sex咖ctio