Music Recommender System

---

Abstract

Music Recommender is to assist users to filter and find out songs consistent with their choice. An ideal recommender will automatically detect preferences and accordingly generate playlists. The content based and collaborative based filtering are integrated to recommend the songs based on tracks, features, genres, and artists. The proposed work is to build a Music Recommender System which can recommend music to listeners based on users’ taste and preferences. It considers into account the list of songs that users heard to or like in their playlist and interacting with devices. KMeans Clustering and BIRCH stands for Balanced Iterative Reducing and Clustering using Hierarchies, were incorporated while building Music Recommender system. The proposed work provides the recommendations based on Artist Wise, Genre Wise, and Mixed recommendations with respect to user choice and preferences. The model gives around 71% accuracy with respect to both KMeans clustering and BIRCH algorithm.

---

Keywords

KMeans, BIRCH Clustering and Recommender System.