The Echo Nest has made a big addition to its Million Song Dataset collection of music metadata. It now includes 1.2 million unique listeners’ taste profiles, overlapping with 400,000 songs in the dataset. The idea is to help researchers understand not just how songs are linked, but what that means for actual listener behaviour.
“With all of this data, we can perhaps begin to better model and understand musical behavior and go far, far beyond ‘people who like Coldplay also like Beyonce’,” explains The Echo Nest’s Brian Whitman in a blog post. He also stresses that the data has been “carefully anonymised” to protect the privacy of the listeners’ whose profiles are part of the dataset.