Behavior analysis and intelligent management decision-making of students in higher vocational colleges based on big data mining

In the rapid development of higher education management, informatization has become an indispensable driving force. With the popularity of the campus card and the accumulation of various business system data, the campus big data environment has initially taken shape. However, there are many shortcomings in the current student information management platform, especially in the accuracy of data mining. To meet this challenge, we use a decision tree, naive Bayes algorithm, and regression analysis to construct a comprehensive analysis model. On this basis, we use Hadoop technology to build a set of behavior analysis and intelligent management system frameworks for higher vocational college students. The system can comprehensively collect and organize students’ study, life and consumption data, and accurately present the overall situation of students on campus. Through in-depth correlation analysis, we can reveal the inner relationship between students’ learning results, life patterns, and psychological changes. To verify the effectiveness of the system, we conducted extensive testing. In the comparison of prediction accuracy of combination algorithms, we used the average relative error as an evaluation index to analyze and predict the behavior of vocational college students. The test data covers a sample of students ranging from 1000 to 6000. The results show that compared with traditional single algorithms, the student behavior prediction model proposed in this study based on multiple algorithm combinations exhibits higher prediction accuracy, with an average relative error of less than 5%. This result indicates that the hybrid algorithm model we constructed has superior predictive performance.