文献类型：期刊文献

英文题名：Source node detection in social networks based on Trust-GMLA algorithm

作者：Zhang, Xueqin[1];Wang, Yan[1];Zhu, Weiyi[1];Liu, Gang[1];Gu, Chunhua[1]

机构：[1]East China Univ Sci & Technol, 130 Meilong Rd, Shanghai 200237, Peoples R China

年份：2022

卷号：189

起止页码：28

外文期刊名：COMPUTER COMMUNICATIONS

收录：;EI(收录号：20221311851172);WOS:【SCI-EXPANDED(收录号:WOS:000806644700004)】；

基金：Funding statement This work was partially supported by the National Natural Science Foundation of China (NSFC No. 51975213) . The funder Chunhua Gu was involved in the manuscript writing.

语种：英文

外文关键词：Social networks; Asynchronous propagation; Source node detection; Trust-GMLA

摘要：In order to solve the problem of source node detection when privacy information leakage occurs in social networks, especially in the case of asynchronous propagation and low forwarding rate, this paper proposes a single source node detection algorithm named Trust-GMLA(Trust-Gradient Maximum Likelihood Algorithm) based on trust network and GMLA(gradient maximum likelihood algorithm). According to the randomness of asynchronous propagation, the hysteresis matrix of GMLA algorithm is reconstructed, and the matrix parameters are automatically adjusted according to the forwarding rate, which improves the success rate of asynchronous diffusion source node detection. Also, aiming at the problem that a large number of repeated paths need to be calculated in GMLA, a trust inference matrix is proposed to infer the trust value between nodes, determine the trust propagation path, reduce unnecessary path calculation, and improve the speed of the algorithm. Experiments on several social networks datasets show that the proposed Trust-GMLA algorithm can effectively improve the source node detection accuracy and speed.