Research And Application Of Trust Mechanism Based On Social Network

With the development of the Internet,Internet service has become an indispensable part of people's life,but there are more serious security issues.In addition to the traditional hard security means,the trust mechanism as an important soft security means has been widely used.Trust relationship has become an important basis for decision-making of Internet users.Therefore,how to establish a reliable trust relationship,or establish a sound mechanism of trust is a hot topic at this stage.This thesis analyzes the existing trust mechanism and fusion emerging special applications-social network,which also belong to the P2 P network topology.We combine the characteristics of P2 P networks and social networks to optimize the trust model,especially to improve the accuracy of trust evaluation,anti-attack ability and algorithm efficiency as the focus of our research,we put forward a series of trust model improvement scheme.The main contributions of this thesis are as follows:First of all,for the trust value evaluation problem,in the selection of trust feature,the context-based multi-influencing factors are adopted to overcome the problem that the traditional trust mechanism is too coarse.Then the gray correlation method is used to adjust the weights of the features dynamically.And the indirect trust calculation based on similarity is also used to solve the situation that the node data is poor.Finally,the simulation results show that the trust model presented in this paper has good trustworthiness and can effectively resist the attacks of malicious nodes.Secondly,the problem of trust feedback in trust mechanism is discussed.In order to adjust the trust value more real-time and dynamic,this paper adopts the dynamic excitation algorithm based on Markov chain prediction.Based on the improvement of the Markov model,the states of the incentive elements based on trust penalty and state transition based on time decay elements are completed.Finally,simulation experiments show that the algorithm has good accuracy in trust forecasting and enhances the defense ability and robustness.Finally,in order to improve the efficiency of the trust mechanism,we propose an algorithm based on node clustering optimization,analyze the objective and interactive attributes of the quantized nodes,construct the logical coordinates,and complete the logical mapping.Then,the KMeans method is used to complete the clustering according to the coordinates of the mapped nodes.This method is used to select nodes.Finally,through simulation experiments,the results show that the node selection of trust model in this chapter not only ensures the accuracy of trust evaluation,but also reduces the time complexity of the algorithm and improves the efficiency of trust mechanism.