Song Recommendation with Non-Negative Matrix Factorization and Graph\n Total Variation

2016/01/08 by Kirell Benzi, Benzi, Kirell, Vassilis Kalofolias +5
Computer Science · #Data Analysis #FOS: Computer and information sciences #FOS: Physical sciences #Information Retrieval (cs.IR) #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Music Technology and Sound Studies #Music and Audio Processing #Recommender Systems and Techniques #Statistics and Probability (physics.data-an)

paper · pdf · doi:10.48550/arxiv.1601.01892

openalex publication_date 2016/01/08 · openalex created_date 2022/10/02 · openalex updated_date 2026/07/28

Abstract

This work formulates a novel song recommender system as a matrix completion\nproblem that benefits from collaborative filtering through Non-negative Matrix\nFactorization (NMF) and content-based filtering via total variation (TV) on\ngraphs. The graphs encode both playlist proximity information and song\nsimilarity, using a rich combination of audio, meta-data and social features.\nAs we demonstrate, our hybrid recommendation system is very versatile and\nincorporates several well-known methods while outperforming them. Particularly,\nwe show on real-world data that our model overcomes w.r.t. two evaluation\nmetrics the recommendation of models solely based on low-rank information,\ngraph-based information or a combination of both.\n

Related

• SeER: An Explainable Deep Learning MIDI-based Hybrid Song Recommender System
• Evaluating Recommender System Algorithms for Generating Local Music Playlists
• Graphs are everywhere -- Psst! In Music Recommendation too
• Evaluating Music Recommendations with Binary Feedback for Multiple Stakeholders
• Item-Item Music Recommendations With Side Information