Personalized Exercise Recommendation driven by Learning Objective within E-Learning Systems

摘要

To enhance the personalization of an e-learning system, an automatic approach of exercise recommendation that is driven by learning objective is proposed. Firstly, the formal models about knowledge points, exercises and their relations are presented based on a course knowledge tree. Then, a computing method is proposed to constantly and automatically update learning objectives in the learning process. According to the learner's learning state, an approach is proposed to accurately describe the learner's learning needs. In order to realize the personalization within the e-learning system, three kinds of influencing factors, including learning objective, the grasp state of knowledge point and learner's answer preferences, are taken into account for the exercises recommendation. A running example is analyzed to demonstrate the feasibility and validity of the proposed approach for recommending exercise to a complete learning objective in a rapid manner.