메타데이터를 이용한 음악 추천 기법

Abstract

Recently, due to the growth of the digital music market, online music service users are having difficulties in selecting their favorite music and wasting a lot of time. In this paper, we propose a recommendation technique to minimize the difficulty of selection and to reduce wasted time. The proposed technique uses an item - based collaborative filtering algorithm that can recommend items without using identifiable personal information. For more accurate recommendation, the listener's preference is predicted by using the metadata of the music and the Top-N music with high preference is finally recommended. We showed that the proposed method improves the performance of the proposed method better than it does when the metadata is not used.