Recommending personalized query revisions

Observing the queries selected by a user, among those suggested by a recommender system, one can infer constraints on the user's utility function, and can avoid suggesting queries that retrieve products with an inferior utility, i.e., dominated queries. In this paper we propose a new efficient technique for the computation of dominated queries. It relies on the system's assumption that the number of possible profiles (or utility functions), of the users it may interact with, is finite. We show that making query suggestions is simplified, and the number of suggestions is strongly reduced. We also found that even if the system is not contemplating the true user profile, among the above mentioned finite set of profiles, its performance is still very close to the optimal one.