Problem recommendation based on point level algorithm

• 1、School of Statistics and Mathematics, Zhejiang Gongshang University, Hangzhou 310018, China
• 2、School of Information and Electronic Engineering, Zhejiang Gongshang University, Hangzhou 310018, China

Abstract：Personalized question recommendation for students is an important research topic in the field of intelligent education. However, the recommendation method of collaborative filtering often neglects the distinction of knowledge point mastery and cognitive level of students. In view of the above problems, this paper proposes a problem recommendation method based on the point level algorithm. The recommended method is divided into three steps. The first step is to construct the weight graph of the hierarchical relations of knowledge points, which can effectively reflect the hierarchical relations among knowledge points. In the second step, according to students\' knowledge of knowledge points, a personalized problem recommendation algorithm is proposed based on the knowledge point hierarchy map. The third step is to obtain the weak knowledge points by updating the student-knowledge point loss rate matrix, so as to realize problem recommendation. Finally, this paper compares the proposed method with the collaborative filtering method, and proves that the proposed method can maintain higher accuracy and interpretability.

Keywords： knowledge points hierarchy map Personalized problem recommendation loss rate matrix