How often and why mode fails to predict mood in low-arousal classical piano music

ABSTRACT

The goals of this paper are the following: (1) To determine the failure rates using Major and Minor as a predictor for positive and negative moods, respectively, (2) To determine the musical features that contributed to making an example a fail case and (3) To determine whether there are significant differences between Natural, Melodic and Harmonic Minor. We addressed these issues by conducting a listening test and manually analysing fail cases among 252 low-arousal classical piano excerpts. We found that Major–Minor categorisation was 90% accurate and high-register Natural Minor excerpts accounted for nearly half of the residual fail cases.

KEYWORDS:

• Music emotion classification
• low-arousal piano music
• mode and register

Acknowledgement

Thanks very much to the anonymous reviewers and editor for their very helpful suggestions that improved the paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Yu Hong  http://orcid.org/0000-0002-8032-4202