Analyzing The Influence Of Social Network Users Combined With The Time Factor

With the rapid development of Internet technology, great changes have taken place in the way of information communication between people. The speed and scale of information dissemination have reached unprecedented levels, especially after the emergence of social networks. People can release their dynamic anytime and anywhere on this platform to generate vast amounts of data every day. Social network plays a more important role in the message communication.Sina Weibo, one of the largest social network platform. People can share informations through the user attention mechanism. Because it spreads the news with real-time characteristics and is not affected by the geographical space, Weibo plays a more and more important role in information dissemination in people’s daily life. Therefore, the study of Sina Weibo contains a huge scientific and commercial value. Nowadays, the research of Weibo users influence is one of the most popular social network issues.Because the behavior of people on weibo is a kind of subjective expression of real life, so this paper proposes a real-time algorithm to compute Weibo users influence based on the dynamics of human behavior theory. The main works of this paper include:First of all, through the research of social network and complex network shows that social network is a kind of complex network, has the basic characteristics of complex networks:small world and scale-free feature. Combined with human behavior dynamics research, At the same time it also discusses the development and research of social networks.Second, by the SINA Weibo API this paper designs the crawling program to crawl experimental datas. Then proved that the time interval of user’s forward behavior is to obey power-law distribution rather than a Poisson distribution combined with human behavior dynamics theories.Then, seted up a mathematical model on user follow behavior and quantified the degree of attention between users of a particular time. This paper Proposed user influence evaluation algorithm based on the thought of PageRank-RTRank. Experiment proves that this algorithm has good convergence and real-time performance.Finally, calculated users influence with the number of followers and PageRank algorithm respectively, Spearman correlation coefficient is used to calculate the correlation between RTRank algorithms and the other methods, analysised the results show that RTRank in calculating the user influence has good real-time performance and comprehensive performance.