Research on the Application of Multidimensional Data Mining Algorithms in Intelligent Management of Higher Education Institutions

Abstract

Based on the campus card data of students at School A's Business School, this paper conducts data mining on student behavior from three dimensions: dining consumption levels, living patterns, and diligence in studying. An improved K-means algorithm is used to classify student behavior data. Then, conduct a correlation analysis of academic performance and behavioral indicators to identify patterns between behavior and academic performance. Use these insights to guide students in improving their poor behavioral habits and enhancing their learning efficiency, thereby improving their academic performance. This paper also improves the Apriori algorithm in association rules and compares it with other algorithms through experiments to verify the feasibility and high efficiency of the improved algorithm. The improved algorithm is then applied to student behavior analysis, revealing the correlation between student behavior habits and academic performance. The main factors influencing academic performance are identified as lifestyle behavior and learning behavior, while consumption behavior has a relatively minor impact on academic performance.