Abstract：Personalized exercise recommendation for students is an important topic in the field of education data mining.In the existing methods of exercise recommendation，most knowledge modeling methods ignore the use of common features among similar students；and collaborative filtering methods ignore the students′ mastery of knowledge.Aiming at the above defects，this paper proposes a personalized problem recommendation method combining deep knowledge tracking model and collaborative filtering method.The method firstly models the student knowledge with the deep knowledge tracking model，and then combines the collaborative filtering method to calculate the correct probability of the student′s exercises，and according to the probability，the exercises within a certain difficulty range are recommended to the students.The method refers to both the knowledge level of the individual student and the similarity information of the students in the similar situation，has better model accuracy，and can recommend suitable content according to the difficulty range.Finally，the effectiveness of the proposed method is verified by experiments.
马骁睿,徐圆,朱群雄. 一种结合深度知识追踪的个性化习题推荐方法[J]. 小型微型计算机系统, 2020, 41(5): 990-995.
MA Xiaorui,XU Yuan,ZHU Qunxiong. Personalized Exercises Recommendation Method Based on Deep Knowledge Tracing. Journal of Chinese Computer Systems, 2020, 41(5): 990-995.
2020, Vol. 41 
