Including Context in a Transactional Recommender System Using a Pre-filtering Approach: Two Real E-commerce Applications

Recent research has shown that including context in a recommender system may improve its performance. The context-based recommendation approaches are classified as pre-filtering, post-filtering and contextual modeling. Moreover, in real e-commerce applications, collecting ratings may be quite difficult. It is possible to use purchasing frequencies instead of ratings, but little research has been done. The research contribution of this work lies in studying when and how including context with a pre-filtering approach improves the performance of a recommender system using transactional data. To this aim, we studied the interaction between homogeneity and sparsity, in several experimental settings. The experiments were done on two databases coming from two actual e-commerce applications.