User Preference Modeling and Exploitation in IoT Scenarios

Recommender Systems are commonly used in web applications to support users in finding items of their interest. We here propose to use Recommender Systems in Internet of Things scenarios to support human decision making in the physical world. For instance, users' choices (visit actions) for points of interests (POIs) while they explore a sensor enabled city can be tracked and used to generate recommendations for not yet visited POIs. In this PhD research, we propose a novel learning approach for generating an explainable human behaviour model and relevant recommendations in sensor enabled areas. Moreover, we propose techniques for simulating user behaviour and analyse the collective dynamics of a population of users.