MusicMate – An Emotion Based Music Suggestion System Via Machine Learning

Authors:

Walpola W.G.D.P.B., Wijerathna M.R.K., Perera R.M, Danoja K.G.S., Tennakoon T.M.L.H.N., Sumanathilaka T.G.D.K.

Keywords:

Emotion, Deep Learning, Machine Learning, Music Recommendation System

Issue Date:

18th February 2022

Abstract:

People ‘s emotions change from time to time. As a result, the majority of people tend to listen to music that reflects their current emotional state. People’s music preferences differ from another and change frequently depending on their emotional state. Although much research has been done in the domain of emotion¬≠based music recommendations, they may not suit all user’s preferences. Identifying key attributes from music which directly affect the human’s mood is unclear and adapting an accurate model for the entire process is challenging. As a solution, this concept paper proposes a music recommendation system which uses numerous machine learning and deep learning techniques based on the user’s emotions, which satisfies their musical preferences. This research focuses on increasing the accuracy of recommendations and satisfies every user.