A Comprehensive Review of Context-Aware Recommender Systems in Education

The purpose of this chapter is to conduct a comprehensive review of context-aware recommender systems. Drawing insights from 21 research papers published between 2017 and 2023, our analysis examines various aspects of these systems in the context of modern education. The primary areas of examination include categorizing contextual factors, methodologies for context acquisition, integration of contextual information, evaluation strategies, and diverse techniques for generating personalized recommendations. The findings of this study are manifold. First, they highlight the widespread adoption of explicit methods for contextual information acquisition. Second, the study emphasizes the extensive use of offline experiments as a strategy for evaluating recommender systems. Third, the findings underscore the multidimensional nature of these systems, demonstrating their capacity to consider factors such as computing context, location, time, activities, and individual user characteristics. Finally, the chapter demonstrates the growing adoption of machine learning techniques in contrast to the comparatively limited use of deep learning techniques within the examined systems. Having highlighted the significance of transparent reporting on methods for incorporating context and stressing the essential requirement for strong privacy and data protection measures, the chapter then advances by identifying promising directions for future research.