Personalized Smart Learning Recommendation System for Arabic Users in Smart Campus

Abstract

The advancement of technologies has modernized learning within smart campuses and has emerged new context through communication between mobile devices. Although there is a revolutionary way to deliver long-term education, a great diversity of learners may have different levels of expertise and cannot be treated in a consistent manner. Nevertheless, multimedia documents recommendation in Arabic language has represented a problem in Natural Language Processing (NLP) due to their richness of features and analysis ambiguities. To tackle the sparsity problem, smart learning recommendation-based approach is proposed for inferring the format of the suitable Arabic document in a contextual situation. Indeed, the user-document interactions are modeled efficiently through deep neural networks architectures. Given the contextual sensor data, the suitable document with the best format is thereafter predicted. The findings suggest that the proposed approach might be effective in improving the learning quality and the collaboration notion in smart learning environment