Research On Information Dissemination And Evolution Model In Social Networks

Mobile social networks have changed the way that people communicate with others,and various social platforms have became the main way for people to seek information.While social platforms provide convenience for us,they also lead many problems.For example,some information,which is negative,malicious or false,exist and spread widely in social networks.There are many reasons for the wide and rapid spread of rumor and the negative information easily.Such as it is difficult to distinguish the authenticity of information because of the huge amount of information on social networks,the threshold of social networks is very low,the cost for information dissemination is very little,and the law of internet supervision is not perfect.The rapid spread of negative or false information in a large area of social networks can easily cause public anxiety and social confusion.Therefore,the state and the government,as well as some enterprises and institutions,are paying more and more attention to the dissemination of information on social networks.This thesis studies the information dissemination on social networks and explores the mechanism of information dissemination.The specific research is as follows:(i)In reality social networks,the user's emotion interacts and spreads with the dissemination of information.Based on this phenomenon,combined with complex network theory,we improved a new SEIR information transmission model on social network based on user's emotion.Firstly,according to the emotional inclination of nodes,the model divides the propagation nodes into positive and negative categories.We consider the influence of emotion on the probability of information propagation,and we established the propagation dynamics differential equation of the model based on the mean-field theory.The equation is shows the rule of information propagation from the perspective of mathematics.Secondly,the proposed model is verified by simulating on the real network and the generated network.Besides,based on the simulations of Weibo network and BA network,the factors affecting information dissemination are studied by contrasting the parameters of the model.Finally,aiming at a hot topic of actual propagation in the real social network,we analyzed its propagation process.The results show that the proposed model could accurately describe the real transmission process of information.(ii)Based on the phenomenon that information spreads in online and offline network simultaneously,we studied the information dissemination and evolution in online and offline coupled networks.Firstly,we summarized the characteristics of online social networks and offline interpersonal networks,and simulated online social networks and offline interpersonal networks with static directed BA network models and dynamic directed WS network models,respectively.And we adopt the assortative link to build a coupled network.Secondly,we explored the evolution mechanism of information propagation on online and offline coupling networks,and analyzed the differences of information propagation evolution between single-layer networks and coupled networks.The experimental results show that information disseminates cooperatively in online and offline coupled networks.Finally,we analyzed the influence of the probability of information interaction between the coupled networks on the information dissemination process,and proved that the probability of online and offline network information interaction is positively correlated with the information dissemination range and speed.(iii)Based on the existing research of information dissemination in social networks,we summarized the rule of information dissemination,and analyzed the impact of both strong and embedded supervisory strategies for the dissemination of information on social networks.Firstly,we theoretically analyzed the reasons for the wide-ranging spread of malicious information.Secondly,combined with the social network topology,we carried out the simulation to verify the spread of malicious information and the effect of the supervisory strategy.Finally,based on the simulation results and theoretical analysis,we proposed a public opinion supervision method on the online and offline coupling network,which provides a theoretical basis for public opinion management or information communication regulatory agencies.