Context-Aware Recommendation of Learning Resources Using Rules Engine

Although numerous studies have been conducted on learning resources recommendation in E-Learning, research extending this investigation into usage of users´ contextual information is rare. This paper presents an innovative architecture of an intelligent personalized context-aware recommendation system in an E-Learning environment. The system offers users by recommending learning materials, tutors, or other learners with common interests combining users´ social tags from external social network and tags about learning materials from autologous collaborative tagging system. A Rules Engine is used to manage a set of rules for each user to achieve personalized recommendation. The rules which define the relationship between tags construct the relationship graph model for users and learning resources thus graph-based collaborative filtering recommendation methods can be implemented. The rules will be adjusted depending on the user´s feedbacks of previous recommendations.