In order to study the influencing factors and the evolution mechanism of network group behavior choice in current society, this paper designs 4 influencing variables, i.e., event fit variable, government influence variable, individual capability variable, and information oriented variable, by applying social physics research methods based on the similarity between particles in dynamics and behavior individual in network society. The event fit variable includes the fit between event sensitivity and public experience. The government influence variable includes communication channels, credibility, and disposition. The individual capability variable includes cognitive ability, mental state, and behavioral state, and the information oriented variable includes information symmetry, impartiality, and authority. We described the attributes of each variable, presented the function of each influencing factor, and established the influence model between different factors. On the basis of the survey of the hot events on the Internet and the statistics of the reports of major media at different stages in the past five years, the paper analyzes the evolution process of network group behavior and verifies the results by Matlab simulation according to the text of “grab salt incident.” Results show that the intensity of fuse incident control directly affects the evolution of secondary events, in which the government plays a leading role. The impartiality and authority of the mainstream media are important factors that influence the evolution of network group behavior. The fit between fuse incident and the public has a certain impact on the network group behavior, but it does not play an absolute role. Individual cognitive ability, psychological endurance, and behavior ability of the public have certain impact on the network group behavior. The rational and objective public who have stronger cognitive ability are less likely to be influenced by others’ opinions.