Key facts
Contents & fields
The dataset includes audio feature files and Billboard data files. The audio features are in a compressed archive msd_audio_features.tar.gz, containing low- and high-level feature JSON files per track, organized by first letter of track identifier. The Billboard data folder contains two CSV files: msd_bb_matches.csv (tracks that appeared in Billboard Hot 100) and msd_bb_non_matches.csv (non-matched tracks as negative samples).
- msd_bb_matches.csv——Contains MSD ID, Echo Nest ID, artist name, track title, release year, peak position in Billboard charts, number of weeks in charts
- msd_bb_non_matches.csv——Contains MSD ID, Echo Nest ID, artist name, track title, release year
- High/low-level feature JSON——Two JSON files per track storing high-level and low-level audio features extracted by Essentia
Research uses
Suitable for music information retrieval, hit song prediction, audio feature analysis, and popular music trend research.
This card was drafted from the source page; institution, coverage, time span, scale, fields and license are subject to the official page (pending human review).
Keywords
Why this is hard to get on your own
When training hit song prediction models, integrating labeled audio features and Billboard chart data from multiple sources is often needed; this dataset provides ready-to-use features and labels.