Managing Emotions in Smart User Models for Recommender Systems

Our research focuses on the development of methodologies that take into account the human factor in user models. There is an obvious link between personality traits and user preferences both being indications of default tendencies in behavior, that can be automated by systems that recommend items to a user. In this work, we define an emotional component for Smart User Models and provide a methodology to build and manage it. The methodology contemplates the acquisition of the emotional component, the use of emotions in a recommendation process and the updating of the Smart User Model according to the recommendation feedback. The methodology is illustrated with a case study.