The Role Of Complex Networks Topology In Opinion Formation Dynamics In The External Field

The purpose of this paper is to study the opinion evolution dynamics on complex network. With computer simulation, we analyze the role of external field and complex network structure in opinion formation.During the last years there have been a great interest in applications of statistical physics ideas and methods in social science. mostly economical models are analyzed using the techniques of stochastic dynamics, percolation theory or the chaos paradigm. Another important subject of this field is the process of opinion formation treated as a consensus phenomena, which is the part of socialphysics.In chapter one, we simply introduce the complexity science concept, complxity system charateristics and the methods of studying complex system. Then, we display the brief history about complex network and opinion dynamics, complex network concepts, complex network models and opinion formation models in chapter two and three.There are many real social systems provided with the small world effect and high clustering coefficient property, which shown in many statistical works. As we know, clustering coefficient not only reflects the probability that your friends are also friends but also reflects how sparse the local society is, which plays an important role in dynamical processes on social network. On the other hand, our society is an open complex system. Around us, the external field, such as the Governentment Policy, the newspaper, the Internet, the TV and other mass medias, plays a special role in shaping the attitudes, beliefs and public opinion. In chaper four, through building our opinion dynamics model, we do research the effects of external field and clustering coefficient (i.e. complex network structure) on opinion formation. With computer simulation, we find that there exists a phase transition from non-consensus phase to consensus phase. In order to researching the role of external field, we define a characteristic quantity t_c, which shows how quick the opinion reach the consensus phase. We find that both of the quantity t_c and the external field magnitude Q meet power law relationship with exponent 0.753±0.023. Futhermore, we analyze the relationship between clustering coefficient C and the order parameter f and find both of them meet double part power law relationship,which shows the role of complex system topology in opinion formation. Finally, we analyze the relationship between the order parameter f and the complex network size N and the clustering coefficient C using finite size scalling effect.