图书情报工作 ›› 2021, Vol. 65 ›› Issue (17): 79-90.DOI: 10.13266/j.issn.0252-3116.2021.17.008

• 情报研究 • 上一篇    下一篇

基于演化事件探测的学科领域科研社群演化特征研究——以图书馆学情报学为例

李纲1, 唐晶1, 毛进1, 田云裴1, 张斌2   

  1. 1. 武汉大学信息资源研究中心 武汉 430072;
    2. 南京大学信息管理学院 南京 210023
  • 收稿日期:2021-03-14 修回日期:2021-06-07 出版日期:2021-09-05 发布日期:2021-09-01
  • 通讯作者: 毛进(ORCID:0000-0001-9572-6709),副教授,硕士生导师,通讯作者,E-mail:danveno@163.com
  • 作者简介:李纲(ORCID:0000-0001-5573-6400),教授,博士生导师;唐晶(ORCID:0000-0002-1211-5812),硕士研究生田云裴(ORCID:0000-0003-0084-6711),硕士研究生;张斌(ORCID:0000-0002-5591-7874),副教授,硕士生导师。
  • 基金资助:
    本文系国家自然科学基金创新研究群体项目"信息资源管理"(项目编号:71921002)和ISTIC-Taylor&Francis Group学术前沿观察联合实验室开放基金(项目编号:IT1910)研究成果之一。

Research on the Evolution Characteristics of Scientific Research Communities in Subject Fields Based on Evolutionary Event Detection-An Example of LIS

Li Gang1, Tang Jing1, Mao Jin1, Tian Yunpei1, Zhang Bin2   

  1. 1. Center for the Studies of Information Resources, Wuhan University, Wuhan 430072;
    2. School of Information Management, Nanjing University, Nanjing 210023
  • Received:2021-03-14 Revised:2021-06-07 Online:2021-09-05 Published:2021-09-01

摘要: [目的/意义] 科研社群作为当代科学研究中重要知识群体,研究其在学科发展过程中的演化特征对于探索领域发展规律、促进知识创新等方面具有重要意义。[方法/过程] 为探究科研社群的动态演化特征,以图书馆学情报学领域为例,从演化事件探测的角度出发,采用Leiden算法对科研社群进行划分,并构建科研社群演化路径与演化树;在此基础上识别科研社群演化事件,从科研社群演化整体分析、科研社群演化路径及演化树特征分析、科研社群演化事件统计特征分析等3个方面来揭示科研社群的演化模式和演化特征。[结果/结论] 研究表明,科研社群规模呈蓬勃发展趋势,科研社群演化树呈现两种演化模式,增长类演化事件大多发生于大型科研社群且发文量较高,新生和消亡演化事件均发生于小型科研社群且发文量较高,合并、部分合并、分裂、衰减等演化事件的平均社群规模较小且其发文量偏低,进一步证明科研社群之间合作交流趋向频繁,科研社群演化日趋复杂。

关键词: 科研社群, Leiden算法, 演化事件, 演化特征

Abstract: [Purpose/significance] Scientific research communities are important knowledge groups in contemporary science. Studying the evolutionary characteristics of scientific research communities is of great significance for exploring the law of field development and promoting knowledge innovation.[Method/process] This article took the field of Library and Information Science (LIS) as an example. From the perspective of evolutionary event detection, this paper used the Leiden algorithm to detect scientific research communities, and constructed their evolution paths and evolution trees. On this basis, this paper identified the evolution events of scientific research communities, and revealed the evolution modes and evolution characteristics of scientific research communities from three aspects:the overall analysis of the evolution, the evolution paths and the characteristics of the evolution trees, and the statistical characteristics of group evolution events.[Result/conclusion] The research shows that the scale of scientific research communities is developing vigorously, and the evolution trees of scientific research communities present two evolution modes. Most of the evolutionary events of growth type occurred in large communities with a relatively high volume of posts, while both ‘form’ and ‘dissovle’ evolution events occurred in small communities with a relatively high volume of posts. The average community size of evolutionary events such as ‘merge’, ‘partial merge’, ‘split’, and ‘shrink’ is small, and the volume of publications is low. These characteristics further prove that the cooperation and exchanges between scientific research communities tend to be frequent, and the evolution of scientific research communities has become increasingly complex.

Key words: scientific research community, Leiden algorithm, evolutionary events, evolutionary characteristics

中图分类号: