In-Course Progressive Prediction and Recommendation for Supporting Personalized Learning

Abstract

Personalized learning is known as an effective educational approach. Individual students’ performance-based recommendations for learning improvement can be useful in supporting personalized learning. In this paper, we propose an in-course fine-grained progressive performance prediction and recommender system that provides recommendations of study topics, materials, and activities, and peers based on predicted grades to help guide individual students for personalized learning within a course. The performance prediction is based on collaborative filtering on the grades of courses and the grades of the course assessments in a course. This in- course prediction and recommendation can be a useful personalized learning supporting tool by continuously and progressively guiding individual students to prepare and do better in the course assessments throughout the entire course.