The impact of negative preferences on a recommendation process

Modern recommender systems rely on user feedback to provide high quality recommendations. User feedback communicates information about the user interests and preferences. However, in most existing recommender systems, only the positive preferences are taken into account in the recommendation process. We are trying, through this paper, to show the impact of negatives preferences on the recommendation process. To do this, we propose a system for recommending movies which combines positive and negative preferences to estimate the utility of a given movie for a given user. Our first experiments show that taking into account the negative preferences has a positive impact on the relevance of the recommendations made by the system.