No abstract

An approach to deal with user preference relations, instead of absolute ratings, in recommender systems is discussed. User's preferences are then ratings expressed qualitatively by using linguistic terms. This is a suitable technique when preferences are imprecise and vague. On the other hand, to avoid that the overall item rating may hide the users' preferences heterogeneity and mislead the system when predicting the items that users are interested in, multi-criteria ratings are used as well. User's items ratings are represented through a preference graph which highlight better items relationships. Similarity between users is performed on the basis of the similarity of their preference relations which can better capture similar users' ratings patterns. Some preliminary results shows that, our approach enhances the classical recommender system precision thanks to the graphs used for prediction which are more informative and reflect user's initial ratings relations in a better way.

No abstract