The algorithm-driven information distribution model fosters the dissemination of information within online communities,and the highly homogeneous communities facilitate the spread of opinions and the contagion of emo-tions.Under the stimulation of sensitive public opinion events,these communities are prone to opinion polarization.Therefore,it is of significant importance to study the phenomenon of algorithm-driven dissemination of public opinion in communities for the governance of public opinion risks.This paper systematically reviews and summarizes the characteristics of public opinion dissemination in communities and identifies the risks associated with this phenomenon.With the application of risk management theory,this paper proposes establishing three major governance systems from the perspectives of risk identification,risk assessment,and risk control:a data-driven and resource-sharing public opinion risk identification and monitoring system,a specialized and normalized risk assess-ment and evaluation system,and a risk governance system with multi-agent participation and efficient collaboration.
algorithmonline public opinionsocial networkingrisk management
NETL
NSTL
维普
万方数据
