知识图谱构建技术综述

KnowledgeGraphConstruction Techniques I，iu Qiao，I，i Yang，DuanHong，LiuYao，andQin Zhiguang (Sf^ooZ o，J竹。内r，玎口fio”dnd So厂r伽rP E竹gi行PPri卵g，U柠iuPrsif_)J o厂EZP(’fr072if SfiP”(、P以门c，TPf^”(JZogy o。厂C^j”n，CAP”gd“ 610054) Abstract Google’s knowledgegraphtechnology hasdrawna lotofresearchattentions in recent years． However，dueto the1imited public disclosureoftechnical details，people find it difficult tounderstand theconnotationandValue of this technology．In this paper，we introducethekey techniques involved in theconstruction of knowledgegraph in a bottom—up way，star“ng from a clearly defined concept and a technicalarchitecture ofthe knowledgegraph．Firstly，we describe in detail the definition and connotati。n of the knowledge graph，and then we propose the technical framework for knowledge graphconstruction，in which the construction process is divided into three levels according to the abstract level of the input knowledge materiaIs，including the information extraction 1ayer，the knowledge iIllegration 1ayer，and the knowledge processing layer，respectively．Secondly，the research status of the key technologies for each 1evel are surVeyed comprehensiVely and also investigated critically for the purposes of gradually revealing themysteries of the knowledge graph technology，the state—of—the art progress，and its relationship with related disciplines．Finally，five major research challenges in this area are summarized，and the correspondingkeyresearch issues are highlighted． Keywords knowledgegraph；semanticWeb；information retrieval；semantic search engine；natural language processin