Research on College Students’ Behavioral Patterns Based on Big Data

With the development of the internet, campus networks and various information systems in colleges have been constructed, and the data generated by accessing the internet can reflect students’ learning and daily behaviors. Traditional methods have difficulty effectively and accurately analyzing the students’ behavior. In this paper, a method is designed to mine students’ behavior patterns. The main contributions of this article include the following: (1) We extract student behavior features from network log data, a total of 44 students’ behavior features were selected in the experiment. (2) We gain 11 online behavior patterns with association rules from students’ daily behaviors. (3) We build a student prediction model based on students’ behavior patterns and the final model predicts the students’ grades with an accuracy of 87.62%. This paper presents a method to analyze students’ behavior patterns. Moreover, reasonable suggestions and effective and credible data support for precise and dynamic decision-making in colleges and universities are provided.