The Spreading Model And Factor Analysis Of Public Opinion Over Complex Social Networks

Since the 1990 s, the rapid development of Internet have has altered the way we live, work and think. Compared with the traditional sociological study of social networks, the complexity of modern social system has greatly increased. In this context, the public opinion spreading also showed many new features, like strong real-time interactivity, rapid spreading, highly informative, wide spreading, many participants etc. That increased the difficulty of management of the relevant departments. In recent years, a spate of public events happened because of rumors spreading, leading to a serious impact on social harmony and stability. Therefore, the master of rules and factors of public opinion in complex social networks is important to offer proper guidance to public opinion and maintain social stability.The whole idea of this paper is to study the process and result of public opinion spreading based on complex network and game theory, grasp the rules and characteristics it, and analyze the influencing factors.On the basis of current research progress in complex social networks and dissemination of public opinion, the interactions between individuals in the social networking groups is modeled from perspective of completely rational game. Based on "perfect rationality", the microscopic interaction behavior between individuals is investigated in a social network. The payoff and cost of public opinion spreading is analyzed, also the trust between individuals, as an important variable, is added to the model. The static game of complete information is established. Then this paper separately analyzed the pure strategy Nash equilibrium and mixed strategy Nash equilibrium of the game. Finally, the simulation experiments showed that trust in social relations between individuals, the importance of public opinion are two important parameters in the game model and was positively correlated with the proportion of public opinion propagation.Then, the interactions between individuals in the social networking groups is modeled from perspective of bounded rational game. And the evolution of public opinion and evolutionary stable strategy of groups is analyzed. Then, we determined the topology model of social networks and strategy update rules of individuals in evolutionary game. Finally, the diffusion process of public opinion was demonstrated by the use of simulation, and several key factors in opinion spreading were analyzed. Experimental results show that the spreading proportion of opinion in scale-free networks and small-world network was significantly higher than that in random networks and regular lattice networks, public opinion in this two networks could spread more widely. At the same time, when investigating the impact of network structure of scale-free networks and small-world networks on public opinion spreading, we found that opinion could spread more widely in scale-free networks, the size of both networks had little influence to the result of the game. Similarly to small-world networks, the network density of scale-free networks are positively correlated with the spreading proportion of opinion. In addition, the intensity of transmission, including the value of spreading and initial distributions of individual perspectives, was positively correlated with the spreading proportion of opinion, and the trust threshold has a significant impact on the result within a certain range. Under the scale-free network structure, when ? ?0.47, the trust threshold was positively correlated with the spreading proportion of opinion, while the limit in the small-world network is almost the same. When the value of the parameter exceeds a certain threshold, the results had no significant difference between the two kinds of networks.