基于传播者心理差异的随机谣言传播模型  

Stochastic Rumor Propagation Model Based on Psychological Difference of Spreaders

作　　者：仓林青 王友国[1] 孙先莉 CANG Lin-qing;WANG You-guo;SUN Xian-li(School of Science,Nanjing University of Posts and Telecommunications,Nanjing 210023,China)

机构地区：[1]南京邮电大学理学院,江苏南京210023

出　　处：《计算机技术与发展》2022年第12期142-149,共8页Computer Technology and Development

基　　金：国家自然科学基金资助项目(62071248);江苏省研究生科研创新计划(KYCX20_0730)。

摘　　要：社交网络已经成为信息时代的主流媒体,因此研究社交网络上的谣言传播动力学行为有助于更好地理解谣言传播机理。考虑到社交网络经常受环境因素的影响,导致网络拓扑结构是时变的,引入随机扰动的噪声,针对一些传播者由于心理差异因素,选择保持沉默,在谣言传播过程中引入沉默者个体,基于此建立一个考虑沉默-复发机制的随机Si-SIR(Silent-Susceptible-Infected-Removed)谣言传播模型,并对复杂社交网络中随机谣言传播动力学方程进行了研究。通过对随机模型进行稳定性分析,推导了谣言消亡的充分条件,讨论了同质网络和异质网络上谣言的传播阈值。通过蒙特卡洛方法分别在WS(Watts-Strogatz)小世界网络、BA(Barabási-Albert)无标度网络、Facebook真实社交网络上进行模拟仿真。仿真实验表明,添加适当强度的噪声,加速了谣言扩散的过程,扩大了谣言的最终规模;与WS和BA网络相比,Facebook网络的谣言扩散速度更快,谣言的最终规模更高;更强的沉默-复发机制能够提高传播节点密度峰值和延迟谣言消亡的时间。Social network has become the mainstream media in the information age.Therefore,studying the rumor propagation dynamics in social network is helpful to better understand the rumor transmission mechanism.Considering that social networks are often affected by environmental factors,resulting in time-varying network topology,random disturbance noise is introduced.Aiming at some spreaders who choose to remain silent due to psychological differences,silent nodes are introduced in the process of rumor spreading.Based on this,a stochastic Si-SIR(Silent-Susceptible-Infected-Removed)rumor propagation model by considering silent-return mechanism is established,and the dynamics equations of stochastic rumor propagation model in complex social networks are studied.Based on the stability analysis of stochastic model,the sufficient condition of rumor extinction is deduced,and the spreading threshold of rumor on homogeneous network and heterogeneous network is discussed.Simulations are conducted in the WS(Watts-Strogatz)small-world network,BA(Barabási-Albert)scale-free network,and Facebook real social network by Monte Carlo method,showing that the rumor diffusion process can be accelerated and expand final scale of rumor as long as add the appropriate noise intensity.The rumor spreading speed of the Facebook network is faster and the higher of final scale of the rumor as long than that in WS network and BA network.Stronger silence-return mechanism can increase the peak of spread nodes density and delay the time of rumor extinction.

关 键 词：随机谣言传播模型 社交网络 噪声 沉默-复发机制 蒙特卡洛方法 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]