Title: Detection and evolution analysis of TCM patent community based on weighted complex network
Authors: Na Deng; Tiansi Du; Xu Chen
Addresses: School of Computer Science, Hubei University of Technology, Wuhan, Hubei, China ' School of Computer Science, Hubei University of Technology, Wuhan, Hubei, China ' School of Information and Safety Engineering, Zhongnan University of Economics and Law, Wuhan, Hubei, China
Abstract: In order to provide help and support for the analysis, mining, research, and development of traditional Chinese medicines, and for the discovery of medicine evolution laws, a weighted complex network based medicine community detection algorithm is proposed. Firstly, a weighted network of herbs is constructed according to the correlation degrees between nodes; and then, nodes with the highest PageRank (PR) value are selected iteratively as the initial clustering centres of the community; lastly, the nodes with the highest similarity to the nodes in the community are added to the community one by one, until all the nodes are divided into the corresponding TCM community. We verify the stability and effectiveness of the algorithm on the classic test networks. The algorithm is applied to the community detection and evolution analysis of antihypertensive TCM patents to obtain the core herbs and their evolution laws.
Keywords: TCM patent; community detection; evolution analysis; weighted complex network; node correlation degree; node importance; node similarity; node direct influence; node common neighbour influence; global modularity.
International Journal of Grid and Utility Computing, 2022 Vol.13 No.6, pp.640 - 651
Received: 14 Mar 2022
Received in revised form: 20 Jul 2022
Accepted: 02 Aug 2022
Published online: 17 Jan 2023 *