E-Learning Course Recommendation Based on Learning Styles and Course Ratings

Abstract

Introduction: Personal learning has become very difficult for effective e-learning systems, which enables the delivery of tailor-made educational content to different students. This article presents an advanced recommendation system that uses Felder-Silverman Learning Style Model (FSLSM) with a hybrid approach that uses both learning-based and course assessment-based recommendations. FSLSM groups users in distinct groups such as active-reflective, sensing-intuitive and visual verbal, allowing the system to identify user preferences through interaction data. The proposed system uses web scraping to gather extensive course details, including multimedia content and user feedback. Learning styles are analysed based on their performance results in the questionnaire. These styles are mapped to the FSLSM model, while course assessments, based on collaborative filtering, improve the recommendation process. By merging these two approaches, the system ensures that the recommended course corresponds to both the user's preferred learning style and the collective quality assessments from other students. Evaluation results show the system's efficiency in increasing the student's satisfaction and commitment compared to traditional recommendation models. This hybrid frame not only provides a dynamic and adaptive learning experience, but also offers a scalable solution for future e-learning platforms.