Modeling and analyzing malware propagation in social networks with heterogeneous infection rates

With the rapid development of social networks, hackers begin to try to spread malware more widely by utilizing various kinds of social networks. Thus, studying malware epidemic dynamics in these networks is becoming a popular subject in the literature. Most of the previous works focus on the effects of factors, such as network topology and user behavior, on malware propagation. Some researchers try to analyze the heterogeneity of infection rates, but the common problem of their works is the factors they mentioned that could affect the heterogeneity are not comprehensive enough. In this paper, focusing on the effects of heterogeneous infection rates, we propose a novel model called HSID (heterogeneous-susceptible–infectious–dormant model) to characterize virus propagation in social networks, in which a connection factor is presented to evaluate the heterogeneous relationships between nodes, and a resistance factor is introduced to represent node’s mutable resistant ability. We analyzed how key parameters in the two factors affect the heterogeneity and then performed simulations to explore the effects in three real-world social networks. The results indicate: heterogeneous relationship could lead to wider diffusion in directed network, and heterogeneous security awareness could lead to wider diffusion in both directed and undirected networks; heterogeneous relationship could restrain the outbreak of malware but heterogeneous initial security awareness would increase the probability; furthermore, the increasing resistibility along with infected times would lead to malware’s disappearance in social networks.