| The rise and vigorous development of new Social Network Service Applications, which mainly involve micro-blog, have gradually changed people’s traditional living habits and social interaction. In micro-blog,"We-Media" users can participate in online social networking through convenient access whenever and wherever possible to get information, exchange ideas and disseminate information. Compared to traditional social network, the new micro-blogging online social network is more flexible and convenient, coupled with the intellectualization of participants, the complexification of social network and the diversification of influence factors, which greatly shortens the time to form, develop and spread Internet public opinions, and then increase the complexity and randomicity of public information’s dissemination and individual opinions’ evolution. Based on this, with the aid of ideas and approaches in cross subjects such as Sociology and Communication Science, this article is to study the micro-blogging network information dissemination mode, multiple topics transmission and competition mechanism, micro individual opinion interaction and macro view emergence process, as well as the node influence identification of online community superposition, and tries to find and restore the macro and micro rules of opinion evolution and information dissemination in micro-blogging network and to establish the corresponding mathematical model, and then, through simulation experiments, to find the rules and ways that can promote or inhibit information dissemination and opinion proliferation. This article will not only help researchers more understand the evolution process of public opinions in micro-blog, but some useful results in the article can offer application extension to the study of complex system dynamics and complex network theories.The research work of this article is greatly supported by the National Natural Science Foundation of China (No.60972012,61172072, and61271308), the Beijing National Natural Science Fund (No.4102047,4112045), the Specialized Research Fund for the Doctoral Program of Higher Education of China (No.20100009110002) and the Beijing Municipal Commission of Education Discipline Construction and Graduate Construction (No. JXKJD20090001). The main work and innovations of this article are showed as follows: 1. Node influence measurement based on online community superposition is established. Excavation of influential node in complex networks has very important value of theoretical research and practical application. The node degree is relatively simple, but it considers less; K-shell’s collapsed node granularity is too coarse; the computational complexity of tightness and betweenness is high in large-scale network, and it’s difficult to guarantee the performance. In addition, community factors have non-ignorable influence on the information dissemination. On these grounds, this article introduces to the community superstition factor and come up with the LWCS Indicators Model, based on the comprehensive analysis of several types of classic node centricity characteristics. The indicator can approximately measure the influence of node transmission and verify the performance of LWCS indicators through the simulation analysis in real network. What’s more, this article introduces to the idea of K-shell partition on the basis of LWCS indicators and come up with the modified indicator LWCS+. The simulation results show that the LWCS+indicator has more powerful advantages in node influence distinguishing and community superposition factors restoration.2. Micro-blogging network information dissemination model--USIR that is based on multidimensional features is researched. Combining with the theories of complex networks, epidemic dynamics and strong and weak relationship, on the basis of considering the multidimensional characteristics of group factors, individual factors (the times of individuals’receiving information and the individuals’ tolerance of information), communication links and so on, it has established the USIR information dissemination model that is more suitable for micro-blog. Through the simulation in real micro-blogging network, it is found that the function of the strong relationship is more obvious than the weak relationship in promoting information dissemination, but removing the weak relationship has a bigger influence on the network topology structure and information dissemination; Network topology has a greater influence on the results of information dissemination. Compared to the scale-free networks and random networks, the information diffusion speed is slower and the dissemination intensity is weaker in the small world network, but its coverage is slightly larger than the scale-free network’s; Individual tolerance has a relatively obvious restrictive function to the intensity and coverage of information dissemination.3. The spontaneous communication process of individuals in micro-blog is studied, and cooperation and competition relationship between information in multiple sources is modeled. Considering the spread of the same topic, multi-source information dissemination model is established, and the different sources of information collaborate with each other. Numerical simulation results show that the individual spontaneous communication process reduces the threshold of the information dissemination, but not significantly expands the scope of information dissemination. Average degree and length of the shortest distance of network can affect information dissemination, but the length of the shortest distance can only play a role when contact infection probability is relatively large. Considering different topics are transmitted at the same time, multi-topic propagation model is established based on users’ interests, and it is a competitive relationship between information sources. Results show that the competition among topics shortens the survival time of the topics, at the same time it helps to produce hot topics. Users’ preference cannot improve the degree of information dissemination in whole situation, but it can increase the participation of a few hot topics. Research results help to explain the rules of the formation of micro-blogging information dissemination and hot topics, providing theoretical basis for guiding strategy research.4. The influence of the network structure and micro-blog users’ trusts on the consensus evolution process is modeled. First of all, the interaction of individual opinions is built, and the opinion evolution processes in different network structures are analyzed. Simulation shows that sufficient information exchange can promote the convergence of group opinions, so as to prevent the polarization and division of macroscopic consensus. The heterogeneity of network topology is beneficial for the minority opinion to obtain the final victory. Then, common evolution model of individual trusts and opinions is set up, and the influence of the individual trusts on the opinion update process is described. The results show that the individual trusts accelerate the consensus evolution process. When the trusts are spread, the opinion cluster is fewer, and the scale of macro opinion cluster is bigger. Increasing the number of strong ties in the network can promote the group opinion formation of crowd followers. However, in order to accelerate the opinion evolution of bigot groups, increasing the heterogeneity of the network is necessary. The results help to understand the function of micro-blogging network and users’ characteristics and to explain the phenomenon of faster formed group opinion. |
PDF Full Text Request