Computer Engineering and Applications ›› 2022, Vol. 58 ›› Issue (22): 116-122.DOI: 10.3778/j.issn.1002-8331.2104-0277

• Pattern Recognition and Artificial Intelligence • Previous Articles     Next Articles

Intelligent Query System for Scientific and Technological Achievements Based on Knowledge Graph

XU Xin, DU Junping, XUE Zhe   

  1. Beijing Key Laboratory of Intelligent Telecommunication Software and Multimedia, School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China
  • Online:2022-11-15 Published:2022-11-15

基于知识图谱的科技成果智能查询系统

徐欣,杜军平,薛哲   

  1. 北京邮电大学 智能通信软件与多媒体北京市重点实验室 计算机学院,北京 100876

Abstract: Since the data of scientific and technological achievements present cross-domain and interdisciplinary characteristics, the traditional information retrieval technology is unable to meet the increasing needs of users for the information acquisition of scientific and technological achievements. This paper analyzes the research status in the field of knowledge graph and information retrieval. The scientific and technological achievements data are crawled from Internet with the web crawlers. Entity recognition and relationship extraction technology is adopted to identify and discover scientific and technological entities, and a knowledge graph of scientific and technological achievements is constructed, so as to realize the structured storage of scientific and technological achievements data. This paper builds an efficient index of scientific and technological entities based on the ElasticSearch, studies the calculation method of semantic similarity of scientific and technological achievements, and realizes an intelligent query system for scientific and technological achievements based on the knowledge graph. The experimental results demonstrate that the constructed system realizes the efficient query of scientific and technological achievements and the association discovery of related contents.

Key words: scientific and technological achievements, knowledge graph, semantic retrieval, intelligent query

摘要: 科技成果数据呈现跨领域、跨学科特性,传统的信息查询检索技术已难以满足用户日益增长的智能化、精准化的科技成果信息获取需求。分析了知识图谱领域和信息检索领域的研究现状。采用网络爬虫从互联网中高效地爬取科技成果数据,利用实体识别和关系抽取技术识别和发现科技成果数据中的科技实体,构建科技成果知识图谱,并实现科技成果数据的结构化存储。基于ElasticSearch搜索引擎对科技实体构建高效索引,研究科技成果语义相似度计算方法,实现基于知识图谱的科技成果智能查询系统。实验结果验证了所构建的系统能够实现科技成果的高效查询以及相关主题内容的关联发现。

关键词: 科技成果, 知识图谱, 语义检索, 智能查询