Research On The Node Importance Measurement For Social Network Rumor Control

In recent years,the rapid development of new media technology has accelerated the large-scale dissemination and sharing of rumors on the Internet,which has had a serious negative impact on society and individuals.Taking effective control measures is an important means to improve the quality of social network ecosystem and create a safe and reliable network environment for people.In the field of social network rumor control,Finding important nodes to have the greatest communication influence has great practical significance for effectively controlling the rumor spread.By introducing the information fusion technology based on D-S evidence theory,this paper systematically studies the node importance measurement of multi feature fusion and the application of important nodes in social network rumor control.The specific research work is as follows:(1)In view of the low accuracy of D-S evidence theory when applied to real data sets,the evidence weight allocation scheme is formulated and a new multiple evidence fusion method based on D-S evidence theory is proposed.In this method,Wasserstein distance formula is used to measure the clarity of evidence,and the credibility of evidence is measured based on jousselme distance and ranking factor.The weight is assigned to the evidence according to the clarity and credibility of evidence,so as to correct the evidence source.Several numerical examples and experiments on real iris data sets show that the proposed method has higher decision accuracy,better convergence and more reliable fusion results than other similar methods.(2)Aiming at the problem that the existing node importance measurement methods can not avoid the impact of network coupling information and information transmission mechanism on the measurement accuracy,a node importance measurement method based on multiple feature fusion(MFF)is proposed.In this method,the improved D-S evidence theory is used to fuse the centrality,transitivity and reputation of nodes,and the nodes are ranked according to the fusion results.Then,experiments are carried out on six real networks to evaluate the proposed method and similar methods from two aspects: robustness,vulnerability and Sir propagation characteristics.Experiments show that the proposed method has better accuracy for node importance measurement,and is more suitable for practical application scenarios of large-scale social networks.(3)Aiming at the selection strategy of rumor refutation nodes in the rumor refutation mechanism in the field of social network rumor control,the proposed MFF node importance measurement method is used to calculate the rumor refutation nodes to be selected.Firstly,the SCDR(suspicious credulous dubious removal)rumor propagation model based on rumor refuting nodes is constructed,and the state transition conditions between nodes are set;Then,the rumor refuting node is determined based on the selection strategy of node importance;Finally,through the simulation of rumor propagation on six real networks,the changes of users who believe in rumor in three scenarios: before control,random node rumor refutation and important node rumor refutation are compared,which proves the superiority of using important nodes as rumor refutation nodes for rumor control,as well as the accuracy and effectiveness of important node measurement algorithm.