Difficulty-Based Knowledge Point Clustering Algorithm Using Students’ Multi-Interactive Behaviors in Online Learning

Abstract

To improve learners’ performance in online learning, a teacher needs to understand the difficulty of knowledge points learners of different cognitive encounter levels in the learning process. This paper proposes a difficulty-based knowledge point clustering algorithm based on collaborative analysis of multi-interactive behaviors. Firstly, combining the group-directed learning path network, forgetting factors and the degree of student-system interaction, we propose a measurement model to calculate the similarity of the difficulty between knowledge points on student-system interactive behavior. Secondly, to solve the data sparsity problem of interaction, we propose an improved similarity model to calculate the similarity of the difficulty between knowledge points on student-teacher and student-student interactive behavior. Finally, the knowledge point difficulty similarity matrix is obtained by integrating the difficulty similarity of knowledge points obtained from student-system interactive behavior, student-teacher interactive behavior, and student-student interactive behavior. The spectral clustering algorithm is used to achieve knowledge point difficulty classification based on the obtained similarity matrix. The experiments on real datasets show that the proposed method has better knowledge point difficulty classification results than the existing methods.