知识图谱 梳理.ppt

知识图谱需要的技术知识图谱架构?知识图谱一般架构:[来源自百度百科]?复旦大学知识图谱架构:?早期知识图谱架构?知识图谱一般架构:[来源自百度百科]架构讨论数据检索预处理构建关系矩阵网络图谱参数调整可视化数据规范化处理结果导读?早期知识图谱架构知识抽取?实体概念抽取?实体概念映射?关系抽取?质量评估KDD2014TutorialonConstructingandMiningWeb-scaleKnowledgeGraphs,NewYork,August24,2014Asamplerofresearchproblems?????????????Growth:plete!Linkprediction:addrelationsOntologymatching:connectgraphsKnowledgeextraction:extractnewentitiesandrelationsfromweb/textValidation:knowledgegraphsarenotalwayscorrect!Entityresolution:mergeduplicateentities,splitwronglymergedonesErrordetection:removefalseassertionsInterface:essknowledge?Semanticparsing:puteanswersusingtheknowledgegraphIntelligence:canAIemergefromknowledgegraphs?AutomaticreasoningandplanningGeneralizationandabstraction972关系抽取定义:常见手段:语义模式匹配[频繁模式抽取,基于密度聚类,基于语义相似性]层次主题模型[弱监督]KDD2014TutorialonConstructingandMiningWeb-scaleKnowledgeGraphs,NewYork,August24,2014Methodsandtechniques???SupervisedmodelsSemi-?Singleentitymethods?????Rule-basedmethodsProbabilisticmodelsFactorizationmethodsEmbeddingmodels80Notinthistutorial:?Entityclassification?Group/expertdetection?Ontologyalignment?:92KDD2014TutorialonConstructingandMiningWeb-scaleKnowledgeGraphs,NewYork,August24,2014?Extractingsemanticrelationsbetweensetsof[grounded]entities?Numerousvariants:?????Undefinedvspre-determinedsetofrelationsBinaryvsn-aryrelations,facetdiscoveryExtractingtemporalinformationSupervision:{fully,un,semi,distant}-supervisionCuesused:onlylexicalvsfulllinguisticfeatures82RelationExtractionKobeBryantLALakersplayForthefranchiseplayerofonceagainsavedmanofthematchfortheLakers”histeam”LosAngeles”“KobeBryant,“Kobe“KobeBryant?102