An algorithm for music recommendation based on the user's musical preferences and desired emotions

Studies show that music is an effective form of emotional induction and can change the emotional behavior of the user. In view of this, this article seeks to create a "selection" algorithm that includes a subset of songs chosen by the user, while taking account of their current emotional state and allowing them to achieve the emotional state that is desired. In carrying out the experiment, we established the current emotional state, the desired emotional state and 20 suggestions of musical items from a sample of 20 users. With the aim of achieving the desired emotional state, the proposed algorithm returned a new list of musical items for each participant. The participants underwent an evaluation during and after playing the music to assess the effectiveness of the returned list. The results suggest that when listening to the songs selected by the algorithm, the user can approach the desired emotional state.