HYBRID SIMILARITY MEASURES FOR MUSIC RECOMMENDATION

This paper describes our submissions for the MIREX 2009 audio music similarity and retrieval task. This task can be regarded as obtaining a suitable distance measurement between songs defined on a certain feature space. First, we propose a high-level semantic measure based on regression by support vector machines of different groups of musical dimensions such as genre and culture, moods and instruments, or rhythm and tempo. Second, we propose a hybrid measure defined as a combination of a number of existing state-of-the-art low-level measures together with the considered semantic measure.