Clustering and Personalized Course Recommendation Based on Improved K-means Algorithm for Student Learning Behavior Data

In order to solve the problem of large individual differences among students and limited accuracy of course recommendation in online learning environments, this paper proposes a personalized course recommendation method based on the improved K-means algorithm. This method introduces a weighted distance measurement mechanism to better distinguish multidimensional learning behavior characteristics; at the same time, a dynamic cluster number adjustment strategy is adopted to adapt to the changes in the heterogeneity and distribution of students' behavior patterns. In the entire system architecture, we first identify students' learning styles through clustering, and then build a behavior-driven recommendation mechanism to achieve a high degree of matching between course content and students' preferences. The experimental part uses data from an English learning platform of a certain university for verification. The results show that the proposed method is superior to traditional methods in both clustering effect and recommendation performance, especially in the two key indicators of Precision@5 and F1-score, while maintaining good stability and generalization ability. This study provides an effective and scalable solution for the application of online education platforms in modeling and recommendation systems, which has strong practical significance.