Research On Information Propagation Model Based On Information Value

In the context of media convergence,compared with the era of traditional media,the information sources and audiences of converged media are wider,and the content distribution speed is faster,leading to some fake and worthless information more easily flooded in the network.It is likely to endanger social stability and even national security if not intervene in time.Therefore,curbing the spread of false and worthless information in the network and creating a good network public opinion environment has become a research focus.Aiming at the problem of fake and worthless information spreading widely on the Internet,this thesis proposes a method to measure information value and network credibility by studying the law of information spreading on social networks.In addition,this thesis constructs an online social network information spreading model,which verifies the accuracy of the model on real data.Finally,the model is applied to an online-offline network,and several different strategies of public opinion control are proposed.The main contents of this thesis are as follows:（1）An online social network propagation model considering known nodes is proposed.In view of the known state in which online social networks may have known information but hesitate to spread it,this thesis proposes a new node state of "known" by using the research method of spreading dynamics and puts forward a communication model of online social networks considering informed nodes.At the same time,the information spreading rules of this model and the traditional SIR model are compared and analyzed on different networks to verify the rationality of this model,which lays a foundation for the study of information value and credibility in the following thesis.（2）A method to measure the value and credibility of information on online social networks is proposed.In order to measure the credibility of information sources,this thesis calculates the credibility of each node through the number of true and fake messages transmitted by each node as a credibility evaluation standard.At the same time,in view of the attribute of information value,due to the difference of value identification and the multi-dimension of value itself,this thesis uses the attractiveness of information on social media to measure the level of information value,and proposes a method to measure information value on social media.The propagation threshold of the model was calculated and verified by mathematical derivation and simulation experiments.Finally,the accuracy of the model was verified by real data on Weibo.（3）Several strategies of public opinion control are proposed and their control effects are analyzed.In view of the problem that users will interact with each other online and offline at the same time in the real world,this thesis constructs an online offline doublelayer network to simulate the social network in the real world.On this basis,different public opinion control strategies are adopted to study their impact on the scope and speed of information spreading,so as to provide theoretical ideas for the control of fake and worthless information in the real world.