Title: Parametrisation and correlation analysis applied to music mood classification

Authors: Bożena Kostek; Magdalena Plewa

Addresses: Gdansk University of Technology, Narutowicza 11/12, 80-233 Gdansk, Poland ' Gdansk University of Technology, Narutowicza 11/12, 80-233 Gdansk, Poland

Abstract: The paper presents a study on music mood categorisation. First, a review of music mood models is presented. Then, the preparation of a set of music excerpts to be used in the experiments and music parametrisation is described. Next, some listening tasks performed to obtain mood descriptors are introduced. Finally, the correlation between mood descriptors and features extracted from parameters is discussed. The paper concludes with some observations derived from the analysis performed.

Keywords: music information retrieval; MIR; music mood models; music emotion; music parametrisation; correlation analysis; mood classification; mood descriptors.

DOI: 10.1504/IJCISTUDIES.2013.054734

International Journal of Computational Intelligence Studies, 2013 Vol.2 No.1, pp.4 - 25

Received: 06 Jun 2012
Accepted: 07 Jul 2012
Published online: 19 Jul 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article