面向科技文献知识元的知识图谱构建研究  

The Construction of Knowledge Graph for Knowledge Elements of Scientific Literature

作　　者：任亮 杜薇薇[1] 刘伟利 REN liang;DU Wei-wei;LIU Wei-li(Institute of Scientific and Technical Information of China,Beijing 100038,China;School of Business and Management,Jilin University,Changchun 130012,China)

机构地区：[1]中国科学技术信息研究所,北京100038 [2]吉林大学商学与管理学院,吉林长春130012

出　　处：《情报科学》2022年第9期26-31,共6页Information Science

摘　　要：【目的/意义】基于知识元挖掘科技文献中的科学知识,建立科学知识之间的联系,构建细粒度知识图谱,旨在打通知识之间的壁垒,满足用户细粒度的知识需求。【方法/过程】首先,本文构建基于图的知识元表示框架,并以知识元为核心构建细粒度知识组织模型;其次,设计面向科技文献知识元的知识图谱,并探究知识图谱构建流程,以实现知识图谱的自动化构建;最后以科技文献中的摘要和引言为实验数据,进行实证研究,构建面向科技文献知识元的知识图谱。【结果/结论】本文所构建的知识图谱不仅能直观地展现学术论文所需要探究的问题、所提出的方法和模型等,还能够揭示科学知识之间的内在关联。【创新/局限】本文探究了细粒度知识组织模型,构建了面向科技文献知识元的知识图谱。在未来的研究中,将不断完善知识图谱构建流程,并探索知识图谱的应用领域。【Purpose/significance】Mining scientific knowledge in scientific literature based on knowledge element, establishing the connection between scientific knowledge, and constructing a fine-grained knowledge graph, aiming to break through the barriers between knowledge and meet the fine-grained knowledge needs of users.【Method/process】Firstly, this paper builds a graph-based knowledge element representation framework, and builds a fine-grained knowledge organization model with knowledge element as the core;Secondly, design a knowledge graph for the knowledge elements of scientific literature, and explore the construction process of knowledge graph, in order to realize automated construction of knowledge graphs;Finally, the abstracts and introductions in scientific literature are used as experimental data to conduct empirical research, and the knowledge graphs for the knowledge elements of scientific literature are constructed.【Result/conclusion】The knowledge graph constructed in this paper can not only intuitively display the questions that need to be explored in academic papers, the proposed methods and models, etc., but also reveal the intrinsic relationship between scientific knowledge.【Innovation/limitation】This paper explores the fine-grained knowledge organization model and builds a knowledge graph for the knowledge elements of scientific literature. In future research, the knowledge graph construction process will be continuously improved, and the application fields of knowledge graph will be explored.

关 键 词：科技文献 知识元 知识图谱 细粒度 知识组织 

分 类 号：G250.2[文化科学—图书馆学]