Research On User Behavior Analysis And Information Diffusion In Social Networks

With the rapid development of internet technology, the social networks are upgrading faster and faster. From the e-mail representing the style of point to point exchange, through the BBS representing the style of point to plane exchange, and then to the activity streams style, a variety of social network platforms come into our life. For its convenience, real-time, grassroots of these factors, microblog has become the main network tool of social networking. Microblog for its users, intricate and complex social networks become a standard. So we can study information diffusion through the user behavior of microblog.At present, the research for social network analysis, focused on the parameters measurement of social networks, structure and content of social network. This paper mainly researched on structure and content of social network, and based on sina weibo, as an example, starting from weibo user behavior. We write a data crawling program to collecting large-scale data through sina microblogging API interface. Through analyzing the data collected, we research weibo on three angles:the microblog user behavioral characteristics, the microblog topology features and microblog information diffusion law and the contact with user behavior。Firstly, for user behavior characteristics, mainly starting from user behavior which can generate interaction, using data collected to analyse the user behavior like forwarding, commenting and praising, it can be concluded that there are a lot of microblog difficult to be concerned. For large number of microblog information, not all of them can be commented. Users always comment on the topic they are interested. And also we find a high volume of microblog forwarded will also get a high number of comments and praise, which is related to the quality of microblog and the time they come out.Secondly, for the study of the characteristics about the topology, the paper added up and calculated these three categories of indicators, like nodes and node degree, clustering coefficient and centrality, the average path length and diameter as a basis for analyzing the topology feature of Sina Weibo.Finally, the articles explored the links such as the law of Sina microblog information diffusion and user behavior, mainly from the microblog being forwarded, the comments and points like to arrive at a point in time of information dissemination affect the speed of information diffusion, information dissemination source quality of influence and impact of information diffusion scale information.In the end,the paper sunmmarized the work and outlook, but also presented from the perspective of the diffusion depth study of information diffusion effect. In addition, it will do a more indepth study.