Research On Social Network Rumor Propagation Model And Traceability Algorithm

Currently,social networks are becoming increasingly connected to people’s daily lives,so people can receive a large amount of information through their social networks.Because the cognitive ability of social network members to know the authenticity of information varies greatly,many people will be deceived by false rumors spread by criminals maliciously through social networks,and at the same time,they may become assistants in the spread of false rumors,which will eventually bring harm to the society.Therefore,by simulating the spread of rumors in real social networks,establishing a rumor spreading model,and combining the status of each node to quickly trace the source of the rumors,is of great significance to suppress the spread of online rumors and ensure the inner circle of social network information.At present,the spreading model of social network rumors is mainly based on infectious disease models such as SI,SIR,SIS,and SEIR,but in essence,the spread of rumors on social networks and the spread of infectious diseases are not the same,and the node status in the spread of rumors is more complicated.Starting from the above background,this paper conducts in-depth research on the modeling and tracing of social network rumors.First,this paper proposes a model based on SIS that can better simulate the spread of rumors in reality.In this model,each network node can include 6 different states.After receiving the rumors,normal users can react quickly through their own knowledge,whether they are infected by the rumors,refute them,or ignore the rumors.In addition,in this model,considering the rumor refuting behavior between real users and the official media,edge dispel rumors nodes and core dispel rumors nodes are added.The model proposed in this paper will enable different rumor-defying nodes to suppress rumors based on the magnitude of the influence the rumors have on the network.In this model,a node that becomes a normal user may still be re-infected,but the probability of re-infection will decrease with the number of infections.The infected node may also become an edge anti-rumor node or a recovery node after receiving the rumorrefuting information.In addition,this model incorporates some real-life communication patterns to improve the authenticity of simulated communication,and establishes a dynamic model of rumor propagation through the node state transition diagram,and finally,through simulation experiments on different complex networks,verifies that the rumor propagation in this model will tend to be stable,thus proving the practicality of this model.Based on the proposed rumor dissemination model,this paper studies the tracing of a single source of rumor dissemination in social networks.Based on the snapshot obtained after the rumor has been spreading for a period of time in the complex network,and on the basis of the classic backpropagation algorithm,the concept of the best transmission path and the eccentricity rate of infection and refutation is proposed,and an improved single-source traceability algorithm for rumor propagation is designed and implemented,through comparative experiments from the accuracy rate and centrality metrics,verifies that the rumor tracing algorithm proposed in this article has a greater accuracy rate.