Music Recommendation System based on Unsupervised Discretization

Because of the revolution in the field of Internet and E-commerce, users are overwhelmed by choices either it may be a book or movie or Music etc. Recommendations systems are serving as one of the important tool to handle information overloading by providing recommendations to users. In this paper we proposed a method to handle music recommendation problem. Unsupervised discretization is used to cluster the items which are similar in their frequency distribution. The proposed method is evaluated by using a benchmark dataset Last.fm. the results depict the fact that the proposed method performs better than the traditional popular recommendation approach.