Explainable Matrix Factorization for Collaborative Filtering

Explanations have been shown to increase the user's trust in recommendations in addition to providing other benefits such as scrutability, which is the ability to verify the validity of recommendations. Most explanation methods are designed for classical neighborhood-based Collaborative Filtering (CF) or rule-based methods. For the state of the art Matrix Factorization (MF) recommender systems, recent explanation methods, require an additional data source, such as item content data, in addition to rating data. In this paper, we address the case where no such additional data is available and propose a new Explainable Matrix Factorization (EMF) technique that computes an accurate top-n recommendation list of items that are explainable. We also introduce new explanation quality metrics, that we call Mean Explainability Precision (MEP) and Mean Explainability Recall (MER).