摘要
近些年,知识图谱的构建技术得到了极大的发展,构建好的知识图谱已经被应用到众多领域。在此基础上,研究者将目光从知识图谱转向事件图谱。事件图谱以事件为核心,准确地描述了事件信息以及事件之间的关联关系。基于此,总结了事件图谱在构建、推理与应用方面的关键技术,主要包括事件抽取、事件信息补全、事件关系推断以及事件预测技术。给出了事件图谱的具体应用场景,并且针对事件图谱研究中存在的挑战,对未来的研究趋势进行了展望。
In recent years,the construction technology of knowledge graphs have been greatly developed,and the constructed knowledge graphs have been applied to many fields.On this basis,the researchers turned their attention from the knowledge graph to the event graph.The event graph takes the event as the core and accurately describes the event information and the relationship between the events.The key technologies of event graphs construction,reasoning and applications were summarized,including event extraction,event information completion,event relationship inference and event prediction.Finally,the specific application scenarios of the event graphs were given,and the future research trends were prospected in view of the challenges existing in the event graph research.
作者
胡志磊
靳小龙
陈剑赟
黄冠利
HU Zhilei;JIN Xiaolong;CHEN Jianyun;HUANG Guanli(Key Laboratory of Network Data Science&Technology,Chinese Academy of Sciences,Beijing 100190,China;Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100190,China;School of Computer Science and Technology,University of Chinese Academy of Sciences,Beijing 100049,China;Beijing Institute of Information Technology,Beijing 100091,China;Beijing Polytechnic,Beijing 100176,China)
出处
《大数据》
2021年第3期80-96,共17页
Big Data Research
基金
国家重点研发计划资助项目(No.2016QY02D0404)
国家自然科学基金资助项目(No.U1911401,No.61772501,No.62002341,No.U1836206)。
关键词
知识图谱
事件图谱
事件知识
事件抽取
事件推理
knowledge graph
event graph
event knowledge
event extraction
event reasoning