|
|
|
| Identifying the Structure of Disciplines through Journal Author Coupling Analysis |
| Lu Xiaoli1,2,3, Wu Dengsheng2 |
1.National Geological Library of China, Beijing 100083
2.Institutes of Science and Development, Chinese Academy of Sciences, Beijing 100190
3.University of Chinese Academy of Sciences, Beijing 100049 |
|
|
|
|
Abstract Calculating the relationship between journals and identifying their disciplinary structure facilitates the exploration of the interaction characteristics between different disciplines and the analysis of the trends of knowledge flow between disciplines. This paper adopted a journal-author coupling method to identify their disciplinary structure. Tackling the problem of information loss due to the traditional use of binary matrix analysis, this study proposed the idea of directly using a journal-author distribution matrix for cluster analysis. However, the high-dimensional data process problem complicated the cluster analysis of the journal-author matrix clustering process. Therefore, this paper proposed a new clustering method using t-SNE dimensionality reduction and a hierarchical clustering model. 69 economics journals from the CSSCI database were selected to be empirically analyzed. A unique dataset, containing 43,617 papers published from 2014-2018 by 47,458 authors, was constructed. The empirical results demonstrated that the t-SNE + hierarchical clustering model proposed in the paper can effectively process and use the journal-author matrix information. It divided the 69 examined economic journals into 9 categories (sub-fields), and clearly specified the types of different categories. The paper also sorted out the premise and applicable conditions of the journal author coupling method to identify the disciplinary structure.
|
|
Received: 16 September 2019
|
|
|
|
1 张琳. 期刊混合聚类的学科分类与交叉学科结构研究[J]. 图书情报工作, 2013, 57(3): 78-84.
2 Wolfram D, Zhao Y H. A comparison of journal similarity across six disciplines using citing discipline analysis[J]. Journal of Informetrics, 2014, 8(4): 840-853.
3 宋歌, 叶继元. 基于SNA的图书情报学期刊互引网络结构分析[J]. 中国图书馆学报, 2009, 35(3): 27-34.
4 张紫琼, 张自立, 叶强, 等. 基于互引网络的管理学部重要期刊内部关系研究[J]. 科学学与科学技术管理, 2010, 31(5): 66-70.
5 岳洪江, 刘思峰. 管理科学期刊同被引网络结构分析[J]. 情报学报, 2008, 27(3): 400-406.
6 王贤文, 刘则渊. 基于共被引率分析的期刊分类研究[J]. 科研管理, 2009, 30(5): 187-195.
7 庞杰, 刘则渊, 姜春林. 基于学科主题共现的国际能源期刊分类与知识结构研究[J]. 科学学与科学技术管理, 2010, 31(9): 15-20.
8 Ni C Q, Ding Y. Journal clustering through interlocking editorship information[J]. Proceedings of the American Society for Information Science and Technology, 2010, 47(1): 1-10.
9 姜春林, 张立伟, 刘盛博. 图书情报学期刊“联锁编委”的社会网络分析[J]. 情报学报, 2014, 33(5): 481-490.
10 张琳, 梁立明, JanssensFrizo, 等. 混合聚类方法用于科学结构研究——以“心理学、社会学和教育学”领域为例[J]. 科学学研究, 2010, 28(6): 837-845, 808.
11 刘玉仙. 基于共期刊学科类间关系构建全学科科学骨架图[J]. 情报学报, 2018, 37(6): 580-589.
12 Ni C Q, Sugimoto C R, Jiang J P. Venue-author-coupling: A measure for identifying disciplines through author communities[J]. Journal of the American Society for Information Science and Technology, 2013, 64(2): 265-279.
13 邱均平, 刘国徽. 基于期刊作者耦合的学科知识聚合研究[J]. 情报杂志, 2014, 33(4): 17-22.
14 孟阳, 吴登生, 李建平. 基于学科布局的科研机构层次聚类模型构建与实证分析[J]. 系统科学与数学, 2019, 39(7): 1054-1067.
15 李琳, 黄海军, 汪寿阳. 基于DEA/AR博弈交叉效率方法的学术期刊评价研究[J]. 管理科学学报, 2016, 19(4): 118-126.
16 吴登生, 李若筠. 中国管理科学领域机构合作的网络结构与演化规律研究[J]. 中国管理科学, 2017, 25(9): 168-177.
17 Jia R X, Nie H H, Xiao W. Power and publications in Chinese academia[J]. Journal of Comparative Economics, 2019, 47(4): 792-805. |
|
|
|
