Analysis and Portrait of College Students' Online Behavior Habits

Abstract

In the era of information technology, the popularity of college networks has brought great convenience to students. However some students indulge in the network and thus affect the quality of learning. In order to help college students make use of network resources properly, we analyze students' surfing behavior from the perspective of data mining. 11 classes of web pages with high degree of attention are taken into consideration. the data of more than 100 million log records generated by students in universities within 20 days are collected. We also use Bisecting K-means algorithm to cluster students on different levels. For measuring every students' behavior, we introduce a stability measure for web surfing behavior. we calculate this parameter for every student and analyse the results. Finally, according to the clustering results, cluster stability analysis results and other attributes, we draw a user portrait for every student. After practice, the method is simple and easy to use, providing reference for universities to organize network management activities and standardizing college students' online behavior.