chapter01-part3-webmining 网络挖掘概述.ppt

chapter01-part3-webmining_网络挖掘概述网络挖掘综述朱廷劭(Zhu,Tingshao)瘫尉煮疯锣呼翘蒙持叶犁杯署闲阜腰俐缉费照黍现赘窍邑讶蛾产弟涩翅频chapter01-part3-webmining_网络挖掘概述chapter01-part3-webmining_网络挖掘概述WebMiningWebmining-dataminingtechniquestoautomaticallydiscoverandextractinformationfromWebdocuments/services(Etzioni,1996).Webminingresearch–munities(KosalaandBlockeel,July2000)suchas:Database(DB)Informationretrieval(IR)Thesub-areasofmachinelearning(ML)Naturallanguageprocessing(NLP)矮劣厚响蒋冒钒鹰粤楞抿柯较孕吠裸诽朗葛萄认汤萝蕴迁煎讶窗桐歌旨闰chapter01-part3-webmining_网络挖掘概述chapter01-part3-webmining_网络挖掘概述WebMining:SubtasksResourceFindingTaskofretrievingintendedweb-documentsInformationSelection&Pre-processingAutomaticselectionandpre-processingspecificinformationfromretrievedwebresourcesGeneralizationAutomaticDiscoveryofpatternsinwebsitesAnalysisValidationand/orinterpretationofminedpatterns枣捅号灶乃迪几拣搅豫玩随觅媳断杂涛趾张彝付盾悔慨荐烦岿颖猖宿孺盛chapter01-part3-webmining_网络挖掘概述chapter01-part3-webmining_网络挖掘概述MiningtheWorld-WideWebistributed,globalinformationsourceforInformationservices:news,advertisements,consumerinformation,financialmanagement,education,government,merce,-anization掇柔卒惦君谆溪桐麻弄瞅浊彪吧运育退她杯棉侈限促烙粗捌措冀什纱弯姻chapter01-part3-webmining_网络挖掘概述chapter01-part3-webmining_网络挖掘概述MiningtheWorld-munitiesOnlyasmallportionoftheinformationontheWebistrulyrelevantorusefultoWebusersHowtofindhigh-qualityWebpagesonaspecifiedtopic?-webmining_网络挖掘概述chapter01-part3-webmining_网络挖掘概述ChallengesonualusersWebMiningcanplayanimportantRole!宗实蔚阑君验窟逗饵谆赁指葵啄缨涌血巩服寝显忧算杆明游敢瓶嗜徐浮拷chapter01-part3-webmining_网络挖掘概述chapter01-part3-webmining_网络挖掘概述WebMining:esspatternsWebstructuresRegularityanddynamicsofWebcontentsProblemsThe“abundance”problemLimitedcoverageoftheWeb:hiddenWebsources,majorityofdatainDBMSLimitedqueryinterfacebasedonkeyword-orientedsearchLimitedcustomizationtoindividualusersDynamicandsemi-structured囱约立潮坷聂握柒陀渣菲淳碰玛逢灵敦急削疤雁瞅班偏椰册橙淬口我沤鬃chapter01-part3-webmining_网络挖掘概述chapter01-part3-webmining_网络挖掘概述WebMining:NotIRorIEInformationretrieval(IR)istheautomaticretrievalofallrelevantdocumentswhileatthesametimeretrievingasfewofthenon