Discerning Influential Spreaders in Complex Networks by Accounting the Spreading Heterogeneity of the Nodes

IEEE Access

Yingchu Xin¹  Xianghua Li¹  Chao Gao¹  Zhen Wang²  Xiyuan Zhen³ 

[1] College of Computer and Information Science, Southwest University, Chongqing, China;School of Mechanical Engineering and Center for OPTical IMagery Analysis and Learning (OPTIMAL), Northwestern Polytechnical University, Xi&x2019;an, China;

关键词: Complex networks;    individual heterogeneity;    network immunization;    propagation;   

DOI  :  10.1109/ACCESS.2019.2927775

来源: DOAJ

【 摘 要 】

Centrality is driven immunization is one of the best ways to prevent massive outbreaks (e.g., rumors and computer viruses) on complex networks, for it can effectively block the important diffusion paths to delay the propagation process. However, most of the previous strategies only consider the topology factor while the individual heterogeneity is widely found in the real world (e.g., different entities have different behaviors and tendencies in transmitting new information) and has an important influence on the propagation process. In this paper, we propose a new heterogeneity oriented centrality that is measured by two heterogeneity factors and one topology factor. First, a heterogeneity factor to describe the frequency of nodes activities (activity rank)is introduced; then a novel conception spread rank is first defined and explained to characterize the spread ability of nodes; finally, one topology factor is combined with the heterogeneity factors. After conducting comprehensive experiments on synthetic and real-world networks by using an interactive email model, the results show that HO centrality could delay the propagation most remarkably than the existing strategies. Therefore, the heterogeneity attributes of nodes should be taken into account when we design a network immunization strategy.

【 授权许可】

Unknown