Meaningful Classification for Music Recommendation Systems
Meaningful Classification for Music Recommendation Systems
Loading...
Issue Date
2021-12
Authors
Geving, Brian
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
While the field of Music Information Retrieval (MIR) is steadily progressing, the majority of music recommendation systems still operate using collaborative filtering. Collaborative filtering operates by finding patterns between music represented in collected playlists. While this method works well for music that has a large amount of representation, it is a very poor method for recommending music that has very little availability. By utilizing machine learning classification algorithms and low-level audio feature extraction, there exists a method to classify music with specific parameters. Instead of using modern music genres as a way to relate music, using labels provided by collaborative filtering can provide a high accuracy way to recommend similar music with low representation.
Description
Keywords
SVM , classification , music , Librosa , Mir , machine learning , collaborative filtering
License
Attribution-NonCommercial-NoDerivs 3.0 United States , openAccess