Research On Modeling And Simulating Of Group Polarization Oriented Public Opinion Evolution

The control of public opinion evolution has always been one of the biggest problems for public opinion regulators.The problem of public opinion evolution is complex and diverse,with the characteristics of strong uncertainty and difficult supervision.Once the supervision is out of place,it always forms one or more severe situations,such as "negative emotional communication","rumors everywhere","serious group polarization phenomenon","rational loss of Internet users".As one of the most serious results of public opinion evolution,the phenomenon of group polarization has always been the focus and difficulty in the field of public opinion evolution research.Especially in today’s society,more and more group polarization events,such as "online Red saya beating pregnant women","Chongqing Bus falling into the river","NBA manager Morey unproper speech" occur in our daily public opinion life.The frequent occurrence of group polarization shows the seriousness of group polarization in the new environment and the public opinion management department urgently needs new and effective supervision methods and guidance theories.In this paper,the public opinion related research theory "self-classification theory" and the new research method "modeling and simulating method" are combined to realize the complementary advantages between the old theoretical research and the new advanced empirical research methods.At the same time,using the advantage of "self-classification theory" in explaining the phenomenon of group polarization,extracting its key social rules,modifying and improving the social theoretical basis of the classic model of public opinion evolution Weisbuch-deffuant model.On this basis,a new public opinion evolution model for group polarization SCT model is proposed.Finally,the computer software is used to simulate the evolution process of the model.The evolution results of public opinion under different network environment,different opinion value distribution,different positive and negative opinion population ratio are summarized and analyzed,and the relevant evolution rules and important factors affecting the formation of group polarization results are obtained.The results show that the average node degree of the network environment has an important influence on the formation of group polarization;the influence of groups with strong prototype mainly lies in the degree and speed of aggregation within the group;different proportion of the opposite groups will lead to the average of views within the majority and the extreme of the minority.Finally,on the basis of the research conclusions,the paper put forward several management suggestions to the government and other relevant public opinion management departments to control the phenomenon of public opinion polarization.