关键词: 在线教育, 试题资源, 推荐系统, 联合概率矩阵分解, 认知诊断

Abstract: In recent years, the number of question resources in online education has grown at an explosive rate. It is difficult for student to find the appropriate questions in the mass of question resources. Therefore, the method of question recommendation for student is proposed to solve this problem. However, many problems exist in traditional question recommendation methods based on unified probalilistic matrix factorization, especially for the fact that lots of information of student knowledge points is not considered, resulting in some questions of low accuracy of recommendation results. Therefore, this paper proposes a kind of personalized question recommendation method based on unified probalilistic matrix factorization. Firstly, this method obtains the informaiton of student knowledge points by the cognitive diagnosis model. Secondly, the processes of unified probalilistic matrix factorization is executed by combining students, questions and knowledge points. Finally, the questions are recommended according to range of difficulty. The experimental results show that the proposed method gets the best recommedation results in the aspect of accuracy of question recommendation in different range of difficulty, compared to other traditional recommendation methods, and has a good application prospect.

Key words: online education, question resource, recommendation system, unified probalilistic matrix factorization, cognitive diagnosis