Research And Application Of Trust Model In Social Network Environment

Along with the rapid development in science & technology and society,social network brings great convenience to people's life,entertainment and work.However,due to the rapid development of social networks,the number of users and the updating speed of information are growing exponentially,which leads to the increasingly prominent trust problem among users in social networks.This paper mainly studies the measurement of trust between users in social networks from two different perspectives.The research contents are as follows.(1)Improved the traditional user node based on social network context information between adjacent users trust measurement model.First,in terms of similarity calculation,refinement of the traditional user similarity calculation.Then,the rationality of the calculated results of the model is improved by combining the similarity of users' basic information and the similarity of users' interest labels.Finally,the effectiveness of the improved model is verified on the user data of tencent weibo.(2)This paper proposes a trust prediction model based on network representation learning,and use the network embedding algorithm in network representation learning to extract the features of nodes in social networks.In Epinions and Ciao data sets compared with other trust prediction model,the experimental results show that the proposed trust model for prediction of further improved its prediction accuracy,and can effectively mitigate its trust data sparse lead to inaccurate prediction of problems,also to a certain extent,improve the real-time and scalability of the trust prediction.(3)In this paper,trust model and collaborative filtering recommendation algorithm are combined to apply to the movie recommendation system.A Django framework based on python is implemented systematically.