作者-主题关联的学科知识网络构建与演化分析
|
摘要
【目的/意义】通过构建作者-主题关联的二模学科知识网络,度量作者在知识创新网络中的影响力,探寻研究活跃程度高、研究范围广、潜在合作空间大的重要学者,对于科学评价学者学术影响力,挖掘热点、前沿研究主题具有重要的指导作用。【方法/过程】基于AT主题模型抽取作者-主题关联矩阵,计算作者的研究主题强度,在此基础上构建作者-主题关联的二模学科知识网络,利用作者在网络中的中心性指标度量作者主题关联影响力;基于复杂网络结构分析方法对学科领域生命周期内作者-主题关联的学科知识网络进行演化分析。【结果/结论】实证分析表明作者主题关联影响力与基于引文的学术影响力和基于社交媒体的社会影响力指标形成有力互补,可用于核心作者以及热点、前沿主题探测。
【Purpose/significance】By constructing the discipline knowledge network based on author-topic association, we can measure the academic influence of authors in this knowledge innovation network. So, the important scholars with high levels of active research, wide research scope, and large potential for cooperation could be found. This is very important for mining hot research topics and research fronts.【Method/process】To construct the discipline knowledge network based on author-topic association, the paper first extracts correlation matrix of authors and topics by AT topic model, then computes research topic intensity of authors. Meanwhile, we put forward method to measure authors' academic influence through centrality index in network. Next, the paper puts forward evaluation analysis method of network structure, by complexity network structure analysis theory.【Result/conclusion】We get important results through empirical analysis. The authors' academic influence based topic association is a powerful supplement to academic influence based on citation and social influence based on social media. Hence, it could be used to find core authors in field of discipline.
【Purpose/significance】By constructing the discipline knowledge network based on author-topic association, we can measure the academic influence of authors in this knowledge innovation network. So, the important scholars with high levels of active research, wide research scope, and large potential for cooperation could be found. This is very important for mining hot research topics and research fronts.【Method/process】To construct the discipline knowledge network based on author-topic association, the paper first extracts correlation matrix of authors and topics by AT topic model, then computes research topic intensity of authors. Meanwhile, we put forward method to measure authors' academic influence through centrality index in network. Next, the paper puts forward evaluation analysis method of network structure, by complexity network structure analysis theory.【Result/conclusion】We get important results through empirical analysis. The authors' academic influence based topic association is a powerful supplement to academic influence based on citation and social influence based on social media. Hence, it could be used to find core authors in field of discipline.
引文
1 Watts D J, Strogatz S H. Collective Dynamics of'Small-World'Networks[J].Nature,1998,393(6684):440-442.
2 Morris S A, Yen G G. Construction of Bipartite and Unipartite Weighted Networks from Collections of Journal Papers[EB/OL].https://arxiv.org/abs/physics/0503061,2015-05-08.
3 吴亚晶,张鹏,狄增如,等.二分网络研究[J].复杂系统与复杂性科学, 2010,7(1):1-12.
4 Hirsch J E. An index to quantify an individual's scientific research output[J]. Proceedings of the National Academy of Sciences of the United States of America,2005, 102(46):16569-16572.
5 Ding Y, Scharffe F, Harth A, et al. AuthorRank:ranking improvement for the web[C]//Proceedings of the2006 International Conference on Semantic Web and Web Services(SWWS’06),2006.
6 王妍,郭舒,张建勇.学者影响力评价指标的相关性研究[J].图书情报工作,2015,59(5):106-112,127.
7 王曰芬,宋爽,苗露.共现分析在知识服务中的应用研究[J].现代图书情报技术,2006,(4):29-34.
8 Borgatti S P, Halgin D S. Analyzing Affiliation Networks[J]. The Sage handbook of social network analysis, 2011,(1):417-433.
9 MannGS,MimnoD,MccallumA.Bibliometric Impact Measures Leveraging Topic Analysis[C]//Proceedings of the 6th Acm/ieee-Cs Joint Conference on Digital Libraries, 2006:65-74.
10 Steyvers M, Smyth P, Rosen-Zvi M, et al. Probabilistic Author-Topic Models for Information Discovery[C]//Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM, 2004:306-315.
11 Blei D M, Ng A Y, Jordan M I. Latent Dirichlet Allocation[J]. Journal of Machine Learning Research,2003,(3):993-1022.
12 李纲,徐健,毛进,周利琴.合著作者研究主题相似性分布研究[J].图书情报工作,2017,(6):92-98.
13 衡晓帆,闫佳丽,汪雪锋,等.基于署名顺序的作者活跃度比较研究[J].情报杂志,2013,32(11):51-54.
14 徐建中,王名扬.文献影响力的综合评价指标体系研究[J].情报理论与实践,2014,37(5):69-72,56.
2 Morris S A, Yen G G. Construction of Bipartite and Unipartite Weighted Networks from Collections of Journal Papers[EB/OL].https://arxiv.org/abs/physics/0503061,2015-05-08.
3 吴亚晶,张鹏,狄增如,等.二分网络研究[J].复杂系统与复杂性科学, 2010,7(1):1-12.
4 Hirsch J E. An index to quantify an individual's scientific research output[J]. Proceedings of the National Academy of Sciences of the United States of America,2005, 102(46):16569-16572.
5 Ding Y, Scharffe F, Harth A, et al. AuthorRank:ranking improvement for the web[C]//Proceedings of the2006 International Conference on Semantic Web and Web Services(SWWS’06),2006.
6 王妍,郭舒,张建勇.学者影响力评价指标的相关性研究[J].图书情报工作,2015,59(5):106-112,127.
7 王曰芬,宋爽,苗露.共现分析在知识服务中的应用研究[J].现代图书情报技术,2006,(4):29-34.
8 Borgatti S P, Halgin D S. Analyzing Affiliation Networks[J]. The Sage handbook of social network analysis, 2011,(1):417-433.
9 MannGS,MimnoD,MccallumA.Bibliometric Impact Measures Leveraging Topic Analysis[C]//Proceedings of the 6th Acm/ieee-Cs Joint Conference on Digital Libraries, 2006:65-74.
10 Steyvers M, Smyth P, Rosen-Zvi M, et al. Probabilistic Author-Topic Models for Information Discovery[C]//Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM, 2004:306-315.
11 Blei D M, Ng A Y, Jordan M I. Latent Dirichlet Allocation[J]. Journal of Machine Learning Research,2003,(3):993-1022.
12 李纲,徐健,毛进,周利琴.合著作者研究主题相似性分布研究[J].图书情报工作,2017,(6):92-98.
13 衡晓帆,闫佳丽,汪雪锋,等.基于署名顺序的作者活跃度比较研究[J].情报杂志,2013,32(11):51-54.
14 徐建中,王名扬.文献影响力的综合评价指标体系研究[J].情报理论与实践,2014,37(5):69-72,56.