Research on rumor purification mechanisms and intervention strategies in social media

There are four common ways to refute rumors on the social media. Namely, professional anti rumor accounts refute rumors, ordinary users report rumors, social networking platforms delete accounts and post system notifications. Quantitative analysis of rumor purification mechanisms can support the implementation of rumor intervention strategies. In this paper, these methods are quantized into the rumor spreading model, mean-field equations are established, the stability analysis and numerical simulations are carried out to analyze the rumor refutation effects. Results show that: the rumor refuting effects by informing ignorants of truth are better than that by persuading rumor spreaders into stopping the rumor spreading. The attendance of individuals knowing the truth in refuting the rumor and its proportion affect the refutation effects. The effects are different when deleting accounts at different rumor spreading periods. The earlier the system notifications published, the better the rumor refuting effects. The enhancement in the reliability and high attendance of the system notifications also have great performance in rumor refuting.