Music Recommendation System using Machine Learning Methods

With the advancement of technology, music has become a widely-consumed entertainment around the world. The existence of such entertainment is a crucial part as an entertainment source which enhances productivity and mood in certain environments. Without it, the world would not be as advanced as it is now with less productivity of workers and scholars, which could lead to unfinished or late work submissions. Recommendation system, which is a result of technology growth, has been implemented for various uses, including music recommender. Due to its popularity, the music recommender system needs to be enhanced to fulfill the demand of music worldwide efficiently with higher accuracy. There are numerous numbers of proposed algorithms implemented for music recommendation systems, such as Random Forest, K-Means, K-Nearest Neighbour, Naïve Bayes, and SVM. This paper is made with a purpose of comparing each machine learning algorithm for music recommendation system to find which of them provides a higher performance, efficiency, and accuracy for a better music recommendation system, with the hopes of contributing to future research on recommendation systems, specifically related to music or any other entertainment sources.