Research on Personalized Recommendation Method of Educational Resources Based on Learner Behavior Analysis

Abstract

The explosion of online learning resources makes the research on personalized recommendation of educational resources increasingly prominent. Based on the theory of learner behavior analysis, this paper analyzes the learning behavior logs of the online learning platform, and constructs a personalized recommendation method for educational resources. It analyzes the learner behavior from the three dimensions of basic attributes, behavior characteristics and result characteristics, and solves the problem of resource analysis. In the simulation process, the method uses the vector space model to complete the modeling of learner behavior, and realizes the division of learner groups based on learner behavior, and further evaluates and optimizes the division of learner groups. In order to further verify the clustering effect in combination with the actual data, the classification results are used for example analysis: the experimental results show that the sample points in the educational resource space are divided into 23 categories according to the distance relationship, and [Formula: see text] is the best. The silhouette coefficient value is 0.54436, and the Calinski–Harabasz score value is 1464.9, which effectively improves the clustering effect of the personalized recommendation method.