Abstract
In the era of information overload, recommender systems (RS) have become crucial tools in improving user experience across various domains, including Learning Management Systems (LMS). RS in LMS is particularly valuable for providing personalized learning recommendations that align with user individual preferences, abilities, and needs. This paper proposes to develop an RS for LMS using transformer models as attention-based systems. By leveraging the self-attention mechanism of transformers, the proposed system can accurately focus on the most relevant aspects of user interactions, resulting in more precise and relevant recommendations. Our experiments compare the transformer-based model with the Neural Collaborative Filtering (NCF) model, demonstrating the superiority of the transformer model in both HR@10 and NDCG@10 metrics. The transformer model achieves HR@10 of 70.59% and NDCG@10 of 50.34%, outperforming the NCF model by capturing more complex interactions between users and learning materials. The results highlight the potential of transformers to enhance personalized learning experiences in LMS, offering a more robust framework for understanding user behavior and delivering tailored learning content.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.