1、合肥工業(yè)大學碩士學位論文基于XML的Web文本挖掘應用研究姓名：馬宏偉申請學位級別：碩士專業(yè)：計算機應用技術(shù)指導教師：胡學鋼20081201ApplicationResearchonWebTextMiningbased—onXMLAbstractWithInternetdevelopmenttheinformationonWWWincreasesfastWWWprovidesmassiveinformationforpeople，but

2、alsocausesUStofallintoacontradictionwhichiSontheonehandpeopleneedtoacquireinformationfromWWWfastandeffectivelyontheotherhand，theinformationonWWWisSOhugestructureoftheinformationiScomplicate，andtherearemanydifficultiesfor

3、dealingwiththeinformationInordertesolvethiscontradiction，theWebminingtechnologyprovidesawayatpresenttheresearchofWebminingisindevelopingstage，andneedsmuchresearchintheory，implementationmethodandtechnologyTheWebminingtech

4、nologyisanapplicationoftraditiondataminingtechnologyundertheⅥ，ebenvironmentwhichiSdiscoveringpatternofimplication，unknown，havinglatentapplicationvalue，uncommonfrommassiveWebdocumentsetandWebinformationofusersbrowsingTheS

5、UbjectiSstudiedinthisdisssertationThemaincontextiSasfollows：ThisdisssertationanalyzestheexistingbasicconceptsandmethodsandtechniquesofdataminingandwebtextminingandXMLThroughresearchingtheprocessingofsemistructureddataand

6、itskeytechniques(egdataextractmethod，transformingarithmeticanddataminingmethodetc)，bringforwardaappliedschemeofXMLbasedWebtextminingFirstly，throughextractingandcleaningandformat—transformingtheinterestedsemistructureddat

7、aonWWW，gainedtheeffectiveXMLformatdataAndfurthermore，useSQLSERVER2005IntegrationServices(SQLSERVERIntegrationServices，，，SSIS’’)andAnalysisServices(SQLSERVERAnalysisServices，”SSAS”)inordertoimplementdatatransformingandloa

8、dingandminingIntheenddesignanddevelopthecompactGUIbyVisualStudioNETandDMXfDataMiningeXtension)inordertobrowseminingresultsbyuserThedissertation’SthoughthasbeencometruethoughdesigningaXML—basedplasticmarketinformationcoll

9、ectionandanalysissystemTheprototypesystempredictsthefuturetrendofplasticmaterialspriceaccordingasitshistorydatacollectingfromWeb，offersaappliedWebminingschemeforextractingandanalyzingandminingtheinteresteddataon艙bKeyword