Design and Development of a song recommender system based on user experience and emotions

Gonzalo J. Osende Pérez. (2019). Design and Development of a song recommender system based on user experience and emotions. Trabajo Fin de Titulación (Graduate Thesis). Universidad Politécnica de Madrid, ETSI Telecomunicación.

Abstract:
This project consist on the making of a music listening program with songs recommendations based in user experience and with an extra condition in the emotion or mood produced by the song. The program will have an interface that allows the user play his song lists or the ones generated by the recommender. In both cases, the user will have the option of filtering the songs by the “emotion” they produce. The recommendation are made based in the listenings the user has done, the more the user listen to a song, the more he like it, and that is the way of getting the actual “taste” of a user. With this, the recommender can guess what song to recommend, comparing it with other user with similar “taste”. All the recommender system will be made by machine learning with the library scikit- learn for Python and other dependencies.