Ensemble Learning in Recommender Systems: Combining Multiple User Interactions for Ranking Personalization

In this paper, we propose a technique that uses multimodal interactions of users to generate a more accurate list of recommendations optimized for the user . Our approach is a response to the actual scenario on the Web which allows users to interact with the content in different ways, and thus, more information about his preferences can be obtained to improve recommendation. The proposal consists of an ensemble learning technique that combines rankings generated by unimodal recommenders based on particular interaction types. By using a combination of different types of feedback from users, we are able to provide better recommendations, as shown by our experimental evaluation.