Hidden influential node selection based on cost-benefit analysis for harmful information propagation control

作     者：Liu, Zhaoli Sun, Qindong Li, Shancang Yang, Zhihai Zhang, Beibei 

作者机构：Xian Univ Technol Sch Comp Sci & Engn Xian Peoples R China Xian Univ Technol Shaanxi Key Lab Network Comp & Secur Xian Peoples R China Univ West England Dept Comp Sci & Creat Technol Bristol England Xian Univ Technol Xian 710048 Peoples R China 

出 版 物：《JOURNAL OF INFORMATION SCIENCE》 (J Inf Sci)

年 卷 期：2025年第51卷第3期

页      面：539-555页

核心收录：

学科分类：1205[管理学-图书情报与档案管理] 12[管理学] 120502[管理学-情报学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：The authors thank Yifei Nan  from Xi'an Jiaotong University for his efforts in problem formulation and data analysis and the anonymous reviewers for their valuable time and comments 

主　　题：Cost-benefit analysis epidemic model greedy algorithm propagation control 

摘      要：The quick development of mobile network and smart devices provides a convenience way for information sharing in online social networks, which also accelerates the propagation of harmful information, thus how to select the hidden influential nodes with lower management cost for reducing the propagation speed of harmful information is an important task. In this article, we propose a greedy hidden influential node selection algorithm based on the epidemic model and cost-benefit analysis. First, we investigate the user behaviour dynamic characteristics from two perspectives of social relationships and interaction behaviours, and then susceptible and infected (SI) epidemic model is applied and user influence is estimated. Second, considering the management cost and benefit of different users, a greedy hidden influential node selection algorithm based on the cost-benefit analysis is proposed. Finally, a series of experiments are conducted using the public social network data set and the data set collected from Sina Weibo, to verify the performance and practicality of the developed method. The experimental results demonstrate that our method outperforms other related methods in harmful information propagation control.