Research On Mechanism Of Opinion Evolution In Complex Network

Along with the progress of networking, the margin between real world and thevirtual is increasingly blurred, as well as the information online and offline.Consequently, the public opinion is disseminated more efficiently than ever, which isalso hard to perceive. To manage with the problem provoked, new method is requiredfor management with regard to its nature and laws. Based on complex network theory,complex adaptive systems (CAS) and opinion dynamics, this paper focused onopinion evolution process in complex network. Within this research, method such asliterature review, empirical study, case study and computer simulation were involvedin relative section. This paper provided in-depth studies of factors of informationcommunication，evolution mechanism of individual opinion and dynamic evolution ofpublic opinion. Contents and conclusions are summarized as follows:This study aimed to explore the mechanism of information communication innews spreading process. Firstly, this paper studied the factors of communicationbehaviors from the perspective of consumer behavior theories and proposed a conceptmodel with relative hypothesis. An empirical validation using585survey results as anexample has been conducted to test the proposed model using a Structural EquationModeling technique. The results of the study show that audience’s attitude, expectedbenefit and perceived risk have strong impacts on their communication decisions.Audience’s trust has strong effects on perceived risk. The presence of informationquality did not strongly influence audience’s trust.For the issue of what are the mechanism and common rules of public opinionevolution in the environment in the presence of new media, this paper studied thefactors from two perspectives, which is information communication and attitudechanging. A theoretical framework describing mechanism was developed. On thisbasis, this paper introduced a complex network consists of585individuals torepresent the social network based on Ising Model and proposed a public opinionevolution model in a weighted and directed complex network using a variation of Glauber Dynamics. A series of Monto Carlo simulations are performed. The resultshows the presence of phase transition in the process of opinion formation andconclusions with practical significance have been provided at the end of this paper.The model in this paper takes local field and social temperature as External variablesto demonstrate attitude evolutions under the influence of Population characteristicsand media. Mechanism of local field and social temperature were discussed. Theresult showed that higher social temperature lead to more critical voice and weakerinfluence of media. In this case, long-term public opinion guiding policy will berequired to ensure the emerging of dominated attitude. On the other hand, lower socialtemperature lead to stronger influence of media.Based on the conclusions above, this paper provided an in-depth study on themechanism of news information communication in complex network from complexadoptive systems perspective. By analyzing the concept model of informationcommunication behavior, this paper proposed a Stimulus-Response Model. By theintegration of communication behavior model and opinion evolution modelmentioned before, a comprehensive simulation framework for public opinionevolution process was proposed. At last, a virtual network has been created to enablemulti-agent simulation performing in specific context. The opinion evolution has beenreproduced and the result shows relative consistency with Realistic data. At the end,several rules and characteristic features of news communication process has beenpresented.