个性化习题路径推荐方法研究综述

A Research Overview of Personalized Exercise Path Recommendation Methods

个性化习题路径推荐技术能够综合考虑学习者的个性化特征,为学习者量身定制习题路径。文章系统地梳理了个性化习题路径推荐研究工作。首先,从推荐方式的角度,介绍了全局最优习题路径推荐和局部迭代习题路径推荐方法,总结了两类推荐方法的优势及其存在的问题。然后,从目前个性化习题路径推荐工作使用较多的核心算法的角度,介绍了基于协同过滤、认知诊断、知识追踪、深度学习和强化学习五类方法。最后,探讨了该领域当前的研究难点,并展望未来研究工作的方向。

The personalized exercise path recommendation technology can take into account the individual characteristics of learners and customize the exercise path for them.In this paper,the research work on personalized exercise path recommendation is systematically reviewed.First of all,from the perspective of recommendation methods,the global optimal exercise path recommendation and local iterative exercise path recommendation methods are introduced,and the advantages and problems of the two types of recommendation methods are summarized.Then,from the perspective of the core algorithms that are widely used in the current personalized exercise path recommendation work,five methods are introduced,which are based on collaborative filtering,cognitive diagnosis,knowledge tracking,in-depth learning and reinforcement learning.Finally,current research difficulties in this field are discussed,and the future research directions are prospected.