Machine Learning

Three billion playlists have been created by 200 million users on spot of fi. Users are grouping tracks for themselves that have some semantic meaning to them, and then they actually label it with a playless name as well. So could you use that information to find that latent embetting? We started playing around with collaborative filtering, and e we sauw tremendous success with it. It would be quite surprising if playlists were actually random, if they had no sementic meaning. For most people, they groupe these tracts for some reason.