Statistical Analysis And Modeling For Weibo User Types And Behaviors

Since 1995,the Internet has made rapid development when we achieved it in our country,and until March 5,2015,Premier Li Keqiang made a report about the government work in the State Council and upgraded the "Internet +" to the national strategic level.And according to the Internet Network Information Center(CNNIC)released the 39 th "China Internet Development Statistics Report" shows that as of December 2016,the scale of Chinese Internet users reached 731 million.In the Web 2.0 era,the participation and activity of Chinese Internet users showed explosive growth,which is not related to the birth of a new media-microblog.At the end of 2016,the sina microblog monthly active users have exceeded 300 million,its fragmentation reading,sharing and interactive services has deeply loved by Internet users.Go to the study of the microblog user's portraits,the influence of ordinary and famous users and the communication process of popular microblog not only helps to understand how information is disseminated and fermented,but also helps people control the spread of things better,understand the Influence of information and the real feelings of Internet users,it has better application value to the public opinion analysis.Through the study,this article has done the following research:First of all,we make use of sina open platform,through the api interface,have the access to get ordinary and famous users' number of follow,number of follower and number of microblog,gender and city data;Through setting the corresponding keywords to get the corresponding hot events' microblog text,events,authentication users,forwarding,comments and other data;Get the corresponding hot events' popular microblog communication process data.Secondly,we make use of the data of famous users and ordinary users to do the crowd portrait analysis.And then make a further study of the correlation between the number of famous users' follow,follower,microblog and the number of ordinary users' follow,follower,microblog and the probability distribution,and give the results of the explanation.Thirdly,do the comment and forward's growth analysis for the popular microblog of the selected hot events,the dissemination of the process analysis and the impact analysis of key users on the event.It reaches the peak of forward and comment when the popular microblog send in half an hour,and the small peak in some spare time period such as 11: 30-13: 30,after 22: 00 and so on.Some influential(famous users)forwarding will also lead to the formation of small peaks,that is,the further dissemination of information.Finally,do text segmentation analysis for the emotions of popular events of the microblog users,extract the higher frequency vocabulary,that is conducive to public opinion analysis.