预览加载中，请您耐心等待几秒...

基于属性网络表示学习的链接预测算法.docx

预览
1/2
2/2

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

基于属性网络表示学习的链接预测算法 Introduction: Linkpredictionisanimportantprobleminnetworkanalysis,whichaimstopredictthelikelihoodoftheexistenceofalinkbetweentwonodesinanetwork.Linkpredictioncanbeappliedtoavarietyofdomains,suchassocialnetworks,biologicalnetworks,andfinancialnetworks.Inrecentyears,representationlearninghasemergedasapowerfultoolforlinkpredictioninnetworks.Thispaperfocusesonattributenetworkrepresentationlearninganditsapplicationtolinkprediction. Background: Inanetwork,eachnodeisassociatedwithasetofattributesthatdescribeitsproperties.Forexample,inasocialnetwork,anodecanbeassociatedwithattributessuchasage,gender,educationlevel,andoccupation.Inabiologicalnetwork,anodecanbeassociatedwithattributessuchasgeneexpressionlevels,proteininteractionpartners,andbiologicalfunction.Theattributesofanodeprovidevaluableinformationaboutitsrelationshipswithothernodesinthenetwork.Therefore,attributenetworkrepresentationlearningaimstolearnalow-dimensionalvectorrepresentationforeachnodeinthenetwork,whichcapturesbothitsstructuralandattributeinformation. Method: Thebasicideaofattributenetworkrepresentationlearningistolearnalow-dimensionalvectorrepresentationforeachnodeinthenetwork,whichcanbeusedforlinkprediction.Therearemanymethodsforattributenetworkrepresentationlearning,suchasnode2vec,DeepWalk,andLINE.Thesemethodsaredesignedtocapturethestructuralinformationofanetwork,suchasnodesimilarityandcommunitystructure.However,theydonotexplicitlymodeltheattributeinformationofnodes. Toaddressthislimitation,researchershaveproposedseveralmethodsforattributenetworkrepresentationlearning,suchasGraRep,TADW,andANRL.Thesemethodsaimtolearnalow-dimensionalvectorrepresentationforeachnodeinthenetworkthatcapturesbothitsstructuralandattributeinformation.GraRepcombinestheadjacencymatrixandattributematrixofanetworkintoahigher-orderadjacencymatrix,whichisdecomposedbysingularvaluedecomposition(SVD)toobtainthenoderepresentations.TADWusesmatrixfactorizationtolearnthejointrepresentationofthenetworkanditsattributematrix.A