Song Recommendation with Non-Negative Matrix Factorization and Graph Total Variation
This work addresses the problem of improving song recommendation accuracy for users, but it appears incremental as it builds on existing hybrid methods.
The paper tackled song recommendation by formulating it as a matrix completion problem, combining collaborative filtering via Non-negative Matrix Factorization and content-based filtering via graph total variation, and demonstrated on real-world data that it outperforms models based solely on low-rank, graph-based, or combined methods in two evaluation metrics.
This work formulates a novel song recommender system as a matrix completion problem that benefits from collaborative filtering through Non-negative Matrix Factorization (NMF) and content-based filtering via total variation (TV) on graphs. The graphs encode both playlist proximity information and song similarity, using a rich combination of audio, meta-data and social features. As we demonstrate, our hybrid recommendation system is very versatile and incorporates several well-known methods while outperforming them. Particularly, we show on real-world data that our model overcomes w.r.t. two evaluation metrics the recommendation of models solely based on low-rank information, graph-based information or a combination of both.