An opinion dynamics model for Online Mass Incident

Online Mass Incident appears with increasing frequency. Its features, e.g. lack of regulation, having a real trigger, make selective attention governed by individual relevance play a significant role in interaction between internet users. Drawing on the research findings of social psychology, we proposed a multi-agent opinion dynamics model for Online Mass Incident, in which selecting likely-minded neighbors or opponents is decided by individual relevance. Numerical simulations show that, high individual relevance favors appearance of large clusters and lessening of extremism, which is qualitatively agree with suggestions of sociologists and actual practice of governments.