A pre-filtering based context-aware recommender system using fuzzy rules

Abstract

Most of recommendation systems work with traditional recommender algorithms such as collaborative filtering. Recently context has been incorporated in these algorithms as a fundamental factor to improve the quality of recommendations provided in the context of the user. Context is used as a filter for selecting items suitable for the current situation of the user. This work contributes to the improvement of contextual recommendations through the proposed method that uses a pre-filtering approach, traditional collaborative filtering and fuzzy rules. The Movie lens dataset was used for testing the method and the experiments conducted show promising results.