The Study Of Student Social Network Based On Fast Unfolding Algorithm

Nowadays,information technology is developing rapidly,and information technology has penetrated into all aspects of life,including college student management systems.Therefore,the campus card management system came into being,and the students' personal information and behavior data are stored in the database.The popularity of the campus digital management system should not only be used to better preserve student data,but also to make these data generate value.By mining valuable information in student data,it provides guidelines for more scientific and efficient student management.In this paper,we used 276 students from a college of Central China Normal University to draw data from more than 20,000 one-card canteens in April 2017 to build a social network and analyze the social situation of 276 students.First of all,by cleaning and preprocessing the data,the consumption location is encoded to facilitate the establishment of subsequent models and data analysis.Then use the sliding time interval method to calculate the number of co-occurrences between students,and delete the false socialization between students by setting the co-occurrence threshold,and then use the Fast Unfolding algorithm to divide the community of students and explore the Social relationship.Comparing the modularity of the two community division results before and after setting the threshold,the comparison results show that after using the co-occurrence threshold,a part of the false social relationship is deleted,and the topological structure of the students 'social networks can be closer to the real social Network conditions.Finally,the traditional community division algorithm GN algorithm and Fast Unfolding algorithm are compared.The results show that the Fast Unfolding algorithm division result is more accurate,and has a larger module value,with better stability and accuracy.Through the division of communities,to understand the overall situation of students'socialization,to determine whether there are student groups and isolation.In addition,this paper analyzes the social situation of the students' personal nodes,which can understand the social situation of each student and the social preferences of the students.Regarding the importance of students' personal nodes,based on the three indicators of degree centrality,intermediary centrality and near centrality,this paper finds students with wide social scope,outliers and "intermediary" students.The relevant work in this article has helped the development of mental health education for students in the School of Chemistry of Huazhong Normal University to a certain extent.By analyzing the relationship of students 'social networks,this article found some outliers,and through the teacher or other students to psychologically Unblocking and helping students get out of difficulties in time.