基于語義的web信息抽取技術(shù)研究

1、Y14370’分類號(hào)：——UDC：——工學(xué)碩士學(xué)位論文密級(jí)：——編號(hào)：——基于語義的Web信息抽取技術(shù)研究碩士研究生：黃文利指導(dǎo)教師：王卓副教授學(xué)位級(jí)別：工學(xué)碩士學(xué)科、專業(yè)：計(jì)算機(jī)軟件與理論所在單位：計(jì)算機(jī)科學(xué)與技術(shù)學(xué)院論文提交日期：2009年1月論文答辯日期：2009年3月學(xué)位授予單位：哈爾濱工程大學(xué)哈爾濱工程大學(xué)碩士學(xué)位論文AbstractAsaglobalinformationspace，Webcontainstremendous

2、intrinsicvalue，howtoextracttheinformationthatuserneedexactlyfromcomplexdatabecomesaveryimportantissueAlthoughagreatdealofresearchhavebeencarriedoutforwebdataextraction，existingtechnologyislackofdescriptiontodataitselfand

3、nevercontainsclearsemanticinformation，pattemisnotspecificneither，whichisdifficulttofittheweb’Scharacteristicofdiversityinstructureandpattem，whichmakesapplicationprogramcannotanalysisandmakeuseofthemassinformationonwebdir

4、ectlywhichcauseshugewasteThisarticleintroducesSuffixTreetechnologycouplingwithdatacharacteristicofaknowledgeintensivewebsite，extractsavailabledatapattem，createsdomainOntologywithProtd96toolsrealizessemanticinformationexp

5、ansionintheprocessofinformationextractionandeliminatestheisomerismofhomogeneousmessagesourceofwebsitebyuseofmethodforestablishingbasedonsemanticThisarticleemphasizestheimplementationtechniqueofinformationdataextractionof

6、knowledgeintensivewebsitebasedontheoverallsolutionofOntologytechnologyandSemi—StructuredwebinformationextractiontechnologyThisarticleputsforwardallOntology—driveninformationextractionpatterninformationpositioningbyfilest

7、ructureandFeatureMatchingviaanalysisandresearchonfundamentalprincipleofconventionalmethodofinformationextraction，technologyanddevelopmentstatus，detailsdesignthinkingofthepatternandtheflowofextractionThissystemresolvesthe

8、isomerismproblemamongthewebfilesbygainingthespecifiedHTMLwebpageinthefirstplace，transferringthewebpagetowellformedXMLfilebasedonthefileConvertingarithmeticbasedonStackStructureandlinkStructure，thenextractingdatapatternfr