Abstract:

Social media is an important channel for people to obtain information at present, but the spreading of misinformation in it has caused social harm, which has puzzled manager. The propagation law of misinformation is studied to reduce the bad social impact caused by misinformation. Misinformation spreading model is established through Multi-Agent modeling method with strong ties social media definition and considering individual heterogeneity and the uncertainty of decision-making behavior. The simulation in various networks is done by using control variable method. The research shows that the initial node degree are not able to completely determine the scope of false information propagation. The length of the mitigating period of the false information dissemination process is related to the network randomness. The distribution of user opinion resonance will affect false information dissemination.

Key words: strong-ties social media, Multi-Agent modeling, D-S theory, simulation

摘要：

社交媒体是当前人们获取信息的重要渠道，但是社交媒体中不实信息大量传播造成社会危害的问题一直困扰着管理者。为了减轻不实信息造成的不良社会影响，对不实信息的传播规律进行研究。划分出强关系型社交媒体，通过Multi-Agent建模的方法，引入用户个体异质性和决策行为不确定性，建立不实信息传播模型。用控制变量法，在多种网络中进行仿真模拟。研究表明：不实信息传播的初始节点不完全能够决定不实信息的传播范围；不实信息传播过程的缓解期长短与网络的随机性有关；用户观点共鸣度分布会对不实信息传播产生影响。

关键词: 强关系型社交媒体, Multi-Agent建模, D-S证据理论, 模拟仿真