Abstract

In the current techno-social environment the importance of providing personalized content for users of social platforms, continually adapting to their needs, is key to the success of the existing platforms. Recommender systems play a crucial role in this regard, but in many cases the level of customization provided is not enough. This article discusses two key aspects related to the development of recommender systems: proactivity and context-awareness. We propose a theoretical reference model for the creation of proactive recommender systems based on context-awareness information. In addition, we check its feasibility in two real-world scenarios where the model has been successfully implemented: banking and personal learning networks. Finally, future courses of action following the contributions proposed are highlighted, with special attention to potential applications in educational environments.