Analysis And Simulation On Group Behavior In The Process Of Network Public Opinion Evolution Based On SOAR Model

At present, the network public opinion has an important influence on the development of social public events. Netizen group is the main body to promote the evolution of network public opinion, and the negative emotion of netizens is one of the most important characteristics of network public opinion. If not properly guided, the negative emotion would likely trigger public crisis. In view of great social influence of the network public opinion, all levels of government pay more and more attention to the management of it, and the research on the network public opinion evolution has become a hot topic. But, existing research lacks study on transformation rules of netizen group behavior, especially in the absence of study on modeling and simulation of transformation rules of netizen group behavior combined with government emergency management. SOAR model uses the concept of "problem space" which is always used in artificial intelligence field, takes cognitive behavior as continuous transformation process of state with time in corresponding problem space, supports modeling and simulation of group behavior transformation rules well.Therefore, based on SOAR model, netizen group is treated as Agent, behavior transformation process of netizen group in network public opinion is considered as continuous transformation process of state with time in the corresponding public opinion space in this paper. Combined with related theories of network public opinion evolution and government emergency management, netizen are divided into Me-formers and In-formers, network public opinion development phase is divided into Generation, Diffusion, Mature and Decline. Then, working memory, long-term memory, decision-making process, learning mechanism of Agent are designed, netizen behavior transformation rule repository is constructed, and SOAR Agent model to reveal behavior transformation rules of netizen group is developed ultimately. Moreover, we designed simulation experiment, select typical public opinion events as samples, use NetLogo as simulation platform, reappearance group behavior transformation process of Weibo users under the influence of different government emergency measures, and validate the validity of SOAR Agent model. Finally, we take the decline phase as an example to assess the effect of different government emergency measures to netizen group behavior of different categories, and put forward suggestions on government emergency management.