last.fm decay ratings
This dataset is addressed to build time-aware music recommender systems when evolution of user preferences is considered. It was built by processing the data collected by Oscar Celma (https://www.upf.edu/web/mtg/lastfm360k) from last.fm. It consists of more than 80,000 songs listened to by 50 users over a 2-year period, creating a collection of more than 420,000 timestamped plays.
/* Style Definitions */
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-bidi-font-family:"Times New Roman";
This dataset is addressed to build time-aware music recommender systems that consider evolution of user preferences over time. The entire dataset was split into a training set containing the data corresponding to the first months and a test set including data from the last 2 months. More information in the Readme file.