摘要
随着社交网络的发展,谣言传播的模拟变得尤为重要。基于传统的D-K谣言传播模型,首先提出一种新的考虑谣言信息和非谣言信息竞争传播的未知者-谣言传播者-非谣言信息传播者-沉默者(IS_rS_nR)模型,并综合考虑网络结构、用户间的交互行为等多种属性以用户群体的平均社会影响力计算不同群体间的转移概率。其次,给出IS_rS_nR模型的平均场方程,并分析方程平衡解的存在性和稳定性。MATLAB的仿真结果显示:(1)非谣言信息的传播与谣言传播的趋势一致;(2)网络平均度、谣言传播者、非谣言信息传播者群体的平均社会影响力对信息传播的传播速度、传播周期及传播范围均有显著作用。研究成果可以在一定程度上对于舆情监测和市场营销起到指导作用。
Modeling rumor spreading is becoming increasingly important,because rumors spread fast and can cause great loss to the society and the economy,especially with the development of online social networks.To solve this problem,an Ignorant-Rumor spreader-Normal spreader-Stifler(IS_rS_nR) model in this paper, which extends the classical Ignorant-Spreader-Stifler,(D-K)rumor spreading model by spiting the spreaders into rumor spreader and normal spreaders and defining the infectious probabilities as functions of user influence.Then,a mean-field equation that describes the dynamics of the IS_rS_nR model is derived.Besides,a steady-state analysis is conducted to analyze the stability of the equilibrium solutions.Furthermore,numerical study is conducted to investigate the final information spreading size under different initial conditions,groups' user influence,decay rates,and average degree of the network.The results based on a numerical study show that(1)the diffusion of non-rumor shows a similar spreading trend as rumor;(2)the number of the initial rumor spreaders can advance the diffusion process;(3)the average degree,the groups' user influence not only can affect the spreading size of the information diffusion,but also can affect the spreading speed and period.Those findings offer a preliminary implication to public opinion monitoring and online marketing.
作者
李志宏
庄云蓓
LI Zhi-hong;ZHUANG Yun-bei(Department of Business Administration, South University of Technology, Guangzhou 510640, China)
出处
《系统工程》
CSSCI
北大核心
2017年第5期62-69,共8页
Systems Engineering
基金
国家自然科学基金资助项目(71571073)
广东省自然科学基金资助项目(2014A030313243)
广东省哲学社会科学十二五规划项目(GD14CGL09)