Title: Research on mining key nodes of complex web-based communities based on mining algorithm
Authors: Yanshan He; Ting Wang; Jianli Xie; Ming Zhang
Addresses: School of Electronic and Information Engineering, Lanzhou Jiaotong University, No. 88, Anning West Road, Anning District, Lanzhou, Gansu 730070, China ' School of Electronic and Information Engineering, Lanzhou Jiaotong University, No. 88, Anning West Road, Anning District, Lanzhou, Gansu 730070, China ' School of Electronic and Information Engineering, Lanzhou Jiaotong University, No. 88, Anning West Road, Anning District, Lanzhou, Gansu 730070, China ' School of Electronic and Information Engineering, Lanzhou Jiaotong University, No. 88, Anning West Road, Anning District, Lanzhou, Gansu 730070, China
Abstract: Mining key nodes in complex web-based communities is of great value for controlling network public opinion and maintaining network safety. This paper introduced two mining algorithms, PageRank and LeaderRank. In order to better realise the mining of nodes, background nodes were introduced to improve LeaderRank, and an improved LeaderRank (ILR) algorithm combining with the average performance of nodes was designed based on the micro-blog web communities. In the case study, taking a micro-blog dataset as an example, PageRank, LeaderRank and ILR algorithms were used to mine key nodes (which has special influence on the structure of function of networks) of micro-blog web community. The results showed that the improved ILR algorithm designed in this paper could combine with the network's topological attributes and average performance and the mined key nodes reflected the number of followers, likes and forwarding, which indicated the influence of nodes better and showed the quality of the mining results from the algorithm. The experimental results prove the reliability of the mining algorithm designed in this paper and provide some theoretical basis for the key node mining work of complex web-based communities such as micro-blog.
Keywords: web-based community; mining algorithm; key node; PageRank LeaderRank.
DOI: 10.1504/IJWBC.2020.107155
International Journal of Web Based Communities, 2020 Vol.16 No.2, pp.202 - 210
Received: 20 Sep 2019
Accepted: 27 Sep 2019
Published online: 04 May 2020 *