Modeling and simulation of public opinion based on multi-agent system

"Public opinion comes before the unrest" has become a consensus.Understanding and capturing the essential mechanism of opinion formation and infection will provide help for forecasting and macroeconomic regulation.Public opinion is an open complex giant system,in which studying opinion based on modeling and simulation analysis is a more comprehensive and effective means.The paper presents one overall model of methodology-method-technology-simulation from the view of systems science besides going through the definitions of Wuli-Shili-Renli and multi-agent system,then summarizes the individuals' behaviors such as "conformity","power","egoism",etc.,analyzes individual choices and constructs three interaction rules.The opinion evolution process was simulated quantitatively and some primary conclusions were obtained.