Improving Melody Classification by Discriminant Feature Extraction and Fusion

Research output: Contribution to conferencePaper


We propose a general approach to discriminant feature extraction and fusion, built on an optimal feature transformation for discriminant analysis [6]. Our experiments indicate that our approach can dramatically reduce the dimensionality of original feature space whilst improving its discriminant power. Our feature fusion method can be carried out in the reduced lower-dimensional subspace, resulting in a further improvement in accuracy. Our experiments concern the classification of music styles based only on the pitch sequence derived from monophonic melodies
Original languageEnglish
Publication statusPublished - Oct 2004
EventISMIR2004, 5th International Conference on Music Information Retrieval - Barcelona, Spain
Duration: 10 Oct 200414 Oct 2004


ConferenceISMIR2004, 5th International Conference on Music Information Retrieval

Cite this