Dynamics of a rumor‐spreading model with diversity of configurations in scale‐free networks

Abstract

SUMMARYIn this paper, we propose a new rumor-spreading model, which is a variant of the susceptible-exposedinfectious-removed epidemic model. We consider the influences of the diversity of configurations as a defense strategy to diminish the damage brought by rumor diffusion. The diversity of configuration can be found in different instant messengers or social networks. Moreover, we apply two kinds of the rumor immunization strategies such as random immunization and targeted immunization and show that targeted immunization is better than random immunization for rumor-spreading control in scale-free networks. We carry out the systematic analysis of the model and study the local and global stability of rumor-free equilibrium. In the proposed model, we introduce two new states, namely, hibernator and stifler2, to study dynamical behaviors of the model. The proposed model combines both targeted immunization and diversity of configuration to prevent rumor spreading. By numerical simulations, we indicate that our model is appreciably better than other models in terms of declining the rumor diffusion speed. Furthermore, we investigate many related dynamic properties to make the result more complete. Simulation results show the effect of different parameters on the rumor diffusion in scale-free networks; meanwhile, numerical simulations confirm the theoretical solution.

A novel organizing scheme of single topic user group based on trust chain model in social network

“…Then, for each topic Θ i ∈ Top, we propose the algorithm of single topic group organizing based on trust chain: [1] topic=Θ, Δ . U = ∅,U C = ∅,U O = ∅; [2] for N i ∈ SNS do [3] if EI N i Θ ð Þ≥μ then (C j = N i ) ∧ (U C ← C j ); [4] for C j ∈ U C do [5] for N i ∈ SNS ∧ N i ∉ U C ∧ ∃ Ω(N i , C j ) do [6] if T(N i , C j ) ≥ η then [7] if [9] p = 0; [10] while p ≠ n do [11] if EI E k Θ ð Þ p ≥ω ∨ MI E k jΘ ð Þ p ≥ω then p = p + 1; [12] else…”

“…Then, impact of topic must be evaluated for selecting influential topics. What is more, information propagation not only is the result of users' behaviors, such as sending, forwarding, or accepting information, but also indicates implicit effects of social relationship based influence diffusion or information spreading in SNS . For instance, it is more probable that information be shared among mutually trusted users.…”

“…What is more, information propagation not only is the result of users 0 behaviors, such as sending, forwarding, or accepting information, but also indicates implicit effects of social relationship based influence diffusion or information spreading in SNS. [8][9][10][11] For instance, it is more probable that information be shared among mutually trusted users. Therefore, user clustering cannot ignore the effectiveness of social relationships in information propagation.…”

Agent-based simulation of the dynamics of malware propagation in scale-free networks

“…In this section, we utilize and modify the rumor spreading model introduced by Hosseini et al 6 to model malware propagation in SFNs. We propose the SEIRS model of malware propagation based on rumor diffusion; we will change the introduced rumor spreading model in Hosseini et al 6 by ignoring the hibernator (represents the nodes that receive the rumor, but they forget it and later can remember it again) and the stifler2 (represents the nodes that never accept the rumor and transmit this rumor again) states in our model, because we do not consider forgetting and remembering mechanisms; furthermore, we assume that the stifler can become susceptible again. Also we consider software diversity and assign diverse software packages to nodes on the network to reduce the outbreak of malware and prevent the exploitability of software vulnerability such as buffer-overflow by a malware.…”

“…In this paper, we consider an agent-based simulation to study the dynamics of malware propagation and investigate the effects of defense mechanisms in SFNs. We introduce a susceptible-exposed-infected-stifler-susceptible (SEIRS) model of malware propagation based on the rumor diffusion model, which is introduced by Hosseini et al 6 From the security viewpoint, the rumor spreading phenomenon is similar to malware propagation, in which all the informed nodes diffuse rumor by informing their neighbor nodes. 7 Rumors can be considered as an ''infection of the mind'' and their diffusion is also similar in nature to an epidemic outbreak.…”