基于知识图谱的问答系统研究与应用  

Research and Application of Question Answering System Based on Knowledge Graph

作　　者：袁博[1,2] 施运梅[1,2] 张乐[1,2] YUAN Bo;SHI Yun-mei;ZHANG Le(Beijing Key Laboratory of Internet Culture Digital Dissemination,Beijing Information Science and Technology University,Beijing 100101,China;School of Computer,Beijing Information Science and Technology University,Beijing 100101,China)

机构地区：[1]北京信息科技大学网络文化与数字传播北京市重点实验室,北京100101 [2]北京信息科技大学计算机学院,北京100101

出　　处：《计算机技术与发展》2021年第10期134-140,共7页Computer Technology and Development

基　　金：国家重点研发计划(2018YFB1004100)。

摘　　要：基于知识图谱的问答系统(knowledge base question answering,KBQA)目前已成为自然语言处理中的热门研究领域。问答系统的应用涉及诸多领域,如医药、电力、交通等各个方面。由此可见,问答系统已成为社会生产发展中必不可少的一项技术。该文聚焦于国内外针对知识图谱的问答系统的研究与应用,对其进行分析梳理,总结了知识图谱、知识库以及问答系统的历史、发展及应用等相关知识,以及现有基于知识图谱的问答系统构建的三类方法,分别为基于模板匹配的方法、基于语义解析的方法以及基于向量建模的方法,探究了深度学习对传统问答系统效果的影响。最后,对基于知识图谱的问答系统技术的未来以及发展进行展望。研究表明:随着人工智能的蓬勃发展,问答系统存在的技术难题不断得到解决,将知识图谱和深度学习技术应用于传统问答系统以提升问答效果已成为大势所趋。At present,KBQA(knowledge base question answering)has become a hot research field in NLP(natural language processing).The application of Q&A(question and answering)system involves many fields,such as medicine,power,transportation and so on.Therefore,Q&A system has become an indispensable technology in the development of social production.We focus on the research and application of Q&A system based on knowledge graph at home and abroad.After analyzing,we summarize the history,development and application of knowledge graph,knowledge base and Q&A system,as well as three kinds of existing methods for constructing Q&A system based on knowledge graph,including the method based on template matching,the method based on semantic parsing and the method based on vector modeling respectively.The influence of deep learning on the effect of traditional Q&A system is explored.Finally,the future development of KBQA is prospected.The research shows that with the vigorous development of artificial intelligence,the technical problems of question answering system will be solved continuously,and it has become a general trend to apply knowledge and deep learning to traditional Q&A system to improve the effect of question answering.

关 键 词：知识图谱 问答系统 模板匹配 语义解析 向量建模 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]