Rumor Suppression Research Based On SDIR Model And Influence Maximizationn Immune Strategy

Online social network makes information spread faster and in a wider range.while the spread of rumor information will bring certain negative effects to people life and economy.Therefore,scholars have carried out researches on rumor propagation models and suppression strategies.At present,rumor propagation models are mostly improved on the basis of infectious disease models,and the lack of consideration of factors such as multi-information competitive dissemination and user behavior characteristics makes rumor propagation not close to reality.However,in the research of strategies to suppress rumors,there are some problems such as ignoring user own influence factors and constructing rumor path is too complicated.In view of the limitations of the above research,this paper mainly studies how to effectively control the rumor spreading in online social networks.The main work is as follows.Firstly,according to the characteristics of emergencies,the influence maximization algorithm UCIM is proposed.Based on the characteristics of user own influence and network topology structure,the algorithm takes into account the number of user followers,user behavior characteristics and clustering coefficient,and combines the activity and clustering coefficient of nodes with their neighbors.Experiments on real data sets verify that UCIM algorithm can obtain the most influential nodes at the current moment better than other algorithms.Secondly,in order to better describe the rumor propagation environment,deny rumor information is introduced on the basis of SIR model,and the SDIR model based on heat and evolutionary game is proposed.The model takes into account the user comprehensive influence,the intimacy between users and the heat influence,and introduces an evolutionary game mechanism;in the evolutionary game mechanism,both internal and external factors are considered,that is,the user’s own discrimination ability and the influence by neighbors;experiments verify that the constructed SDIR model constructed can be closer to the reality,and the deny rumor information introduced into the model can effectively suppress the rumor information diffusion and dissemination to a certain extent.Finally,in order to effectively suppress the spread of rumor information,IS-APWUC immune strategy based on two-stage influence maximization is proposed.The first stage is to use UCIM algorithm to mine the key role of public opinion leader nodes;The second stage is to use the existing rumor detection method to identify the rumor node for the leader node,and to prune the weak neighbors with the rumor set as root to build a simplified rumor path tree.And considering the two factors of infection by rumor nodes and comprehensive influence,the nodes with the greatest influence are found to be the immune nodes.Through simulation experiments in SIR model and SDIR model,it is verified that IS-APWUC algorithm has a smaller rumor propagation range compared with other algorithms,and achieves the effect of suppressing rumor propagation.