Million Song Dataset

Dataset of information derived from and related to one million contemporary songs, with more than 50 variables (including information on track metadata, social networks, and more)

Type:

Dataset

Applicable Fields:

music, tagging, categorization, language, network analysis, language use, group behavior, exploration, perception

Access Requirements:

Free

Example Publications:

Bertin-Mahieux, Ellis, Whitman, & Lamer (2011). The Million Song Dataset. Proceedings of the 12th International Society for Music Information Retrieval Conference.

URL:

http://labrosa.ee.columbia.edu/millionsong/pages/getting-dataset