Solely Tag-Based Music Genre Classification

As a fundamental and critical component of music information retrieval (MIR) systems, automatically classifying music by genre is a challenging problem. The approaches depending on low-level audio features may not be able to obtain satisfactory results. In recent years, the social tags have emerged as an important way to provide information about resources on the Web. In this paper we are interested in another aspect, namely how perform automatic music genre classification solely depending on the available tag data. Two classification methods based on the social tags (including music-tag and artist-tag) which crawled from Last. fm are developed in our work. The first one, we use the generative probabilistic model Latent Dirichlet Allocation (LDA) to analyze the music-tag. Then, we can compute the probability of every tag belonging to each music genre. The starting point of the second method is that music's artist is often associated with music genres more closely. Therefore, we can calculate the similarity between the artist-tags to infer which genre the music belongs to. At last, our experimental results demonstrate the benefit of using tags for accurate music genre classification.