基于生成對抗網(wǎng)絡(luò)與漸進(jìn)式融合的多模態(tài)實(shí)體對齊

打開文本圖片集

Multimodal entity alignment based on dual-generator shared-adversarial network

Feng Guanga?，Zheng Runting?，Liu Tianxiang?，Yang Yanru?，Lin Jianzhonga， Zhong Tinga，HuangRongcan?，XiangFeng?，LiWeichenb （a.SchoolofAutomation，b.SchoolofComputerScience，Guangdong UniversityofTechnology，Guangzhou510o6，China）

Abstract：Inthefieldofeducation，knowledgegraph fusionplaysacrucialrole.Asacore technologyof knowledge graphfusion，entityalignmentaistoidentifyequivalent entitypairsacrossmultiple knowledge graphs.Most existing entityalignment methodsassume thateachsourceentityhasacorresponding entityinthetargetknowledge graph.However，whenusingcrosslingualandcros-raphetitysets，theproblemofdanglingentitiesarises.Toaddresstisissue，thispaperproposedthedualgeneratorshared-adversarial network entityalignment model（DGSAN-EA）.This modelutilized partialparametersharig and anoptimalselectionstrategytotraintwogenerators，selectingtheoptimalgneatortoconditionallgenerateewetisacoss knowledgegraphs，therebyenhancing thedatasetand solving thedangling entityproblem.Furthermore，aprogressive fusion strategyandtheintroductionofdistributionconsistencylossfunctionefectielyresolvethedistortionoffusedfeatureformationandthemisalignmentbetweenmodalitiesinmultimodalentityalignment.Validationonmultiplepublicdatasetsshows that compared to existing multimodal entityalignment models，DGSAN-EAachieveshigher hit@ kand MMRscores，demonstrating itseffectiveness in entity alignment tasks.

Key words：knowledge graph（KG）;entity alignment;adversarial network；dual generator;parameter sharing；progresive fusion；distribution consistency

0 引言

在大數(shù)據(jù)時(shí)代背景下，知識圖譜（KG）作為結(jié)構(gòu)化知識表示的重要工具，其應(yīng)用已擴(kuò)展至教育、醫(yī)療、金融等關(guān)鍵領(lǐng)域。（剩余22678字）