P2P Music Recommender System Based On a Single-Scaled Hybrid Filtering

This paper proposes a new P2P-based hybrid music recommendation system for internet audio applications. The proposed system combines the content-based filtering (CBF) and the collaborative filtering (CF) algorithms. A unified scale allows direct comparison between the scores generated from both algorithms, thus recommendation can be made based on the scores without regarding to which algorithm each selection comes from. Experimental results show that the proposed system yields a better and more consistent recommendation hit ratios.