Applying the HYRIWYG incentive mechanism in a recommender system

Abstract

The use of recommender systems (RS) has become common in several e-commerce Web sites. Collaborative filtering is one of the most popular RS techniques. It infers a user's predilections based on similarity with other users. However, to work correctly, the RS depends on the evaluator's participation with truly opinions. The HYRIWYG mechanism proposes an incentive model to motivate users to contribute with a high number of opinions and, more important, guaranteeing the truthfulness of the information. In this paper, we present HYRIWYG mechanism and discuss it using an empirical study in the movie evaluation domain. Initial results highlight the great potential benefits of using this kind of incentive mechanism.