Music Recommendation Model Using Machine Lear...

Music Recommendation Model Using Machine Learning

Author Name : K Madhu Chnadra Prasad, Dr Nachappa MN

ABSTRACT

The music recommender can predict and then offer appropriate songs to their users using a music recommender system based on the characteristics of previous music heard. We built, constructed, and assessed a music recommendation system for this project. We utilized Kaggle's Million Song Dataset to uncover connections between users and songs, as well as to learn from users' prior listening behavior to generate suggestions for songs they would want to listen to the most. We will detail the methodologies we used, as well as the outcomes and their analysis, in this paper. For the collaborative filtering algorithm, we achieved the best results. We've also created a popularity-based model that follows the trend. Music recommendation can be a difficult problem because we have to structure music in such a way that we can recommend songs to clients who don't have a specific preference. It's fluid, and it's sometimes influenced by factors other than a client's or song's listening history.

Key Words: recommendation systems, music, Million Song Dataset, collaborative filtering, content-based, popularity-based, Machine Learning (ML).