A Study Of Analysis And Detection Of College Students On Weibo

In the information age, social networks have changed people’s life a lot. Weiboare very popular in college students.These young are willing to try new things,andare influenced by weibo greatly. The purpose of this paper is to analysis the featureof college students’ behavior in weibo, construct a classifier to detect collegestudents in weibo, and make a suggestion for university management. This paperselects3000users from the Sina micro-blog randomly, then analyzed their data bythe method of statistical analysis and did some experiments. The main contents ofthis paper include:First of all, analyze the university student users’ feature in weibo. This paperrandomly selected and manually labelled1502college students users and1498nonstudent users, and collected their basic information and weibo content. Based on theanalysis of these data, this paper summarized the characteristic of the way collegestudent users used weibo.Secondly, a method to identified college students in weibo. The paper maked3000users as experimental samples. And a classifier was constructed to detectcollege students in weibo, with keywords in weibo and educational institutions asfeatures. The experimental results showed that, machine learning can be used for theidentification of student users.Finally, according to related research, and the results in this paper, we putforward some measures for student management in our colleges in the environmentof social network. Colleges should improve the management system, establish asupervision system and strengthen students’ education.