Dimensionality Reduction and Classification Analysis on the Audio Section of the SEMAINE Database

Authors

Ricardo A. Calix, Louisiana State University
Mehdi A. Khazaeli, Louisiana State University
Leili Javadpour, University of the PacificFollow
Gerald M. Knapp, Louisiana State University

ORCiD

Leili Javadpour: 0000-0003-4004-1950

Document Type

Conference Presentation

Conference Title

Humaine Association Conference on Affective Computing and Intelligent Interaction

Location

Memphis, TN

Conference Dates

October 9-12, 2011

Date of Presentation

10-12-2011

ISSN

0302-9743

Issue

2

DOI

10.1007/978-3-642-24571-8_43

First Page

323

Last Page

331

Abstract

This paper presents an analysis of the audio section of the SEMAINE database for affect detection. Chi-square and principal component analysis techniques are used to reduce the dimensionality of the audio datasets. After dimensionality reduction, different classification techniques are used to perform emotion classification at the word level. Additionally, for unbalanced training sets, class re-sampling is performed to improve the model's classification results. Overall, the final results indicate that Support Vector Machines (SVM) performed best for all data sets. Results show promise for the SEMAINE database as an interesting corpus to study affect detection. © 2011 Springer-Verlag.

Recommended Citation

Calix, R. A., Khazaeli, M. A., Javadpour, L., & Knapp, G. M. (2011). Dimensionality Reduction and Classification Analysis on the Audio Section of the SEMAINE Database. Paper presented at Humaine Association Conference on Affective Computing and Intelligent Interaction in Memphis, TN.
https://scholarlycommons.pacific.edu/esob-facpres/387