Improving recommender systems by increasing choice satisfaction through personalization: a case study of Amazon’s and Netflix’ recommender systems

Abstract

This research aims to increase the efficiency of the recommender system, therefore increasing choice satisfaction. A case study on the recommender systems of Amazon and Netflix alluded a proposal which states to increase transparency of the given recommendations to the user through additional user preference settings of which they are shown their impact on the ranking of the list of given recommendations. These indications lead to a larger understanding of recommender systems by the user, which increases their quality of feedback and choice satisfaction, thus improving the efficiency of the recommender system. Furthermore, it discusses open problems that would be resolved with the implementation of this proposal, wider patterns recognized in other recommenders and the future of recommender systems.