|
Published Articles >> Table of Contents >> Abstract
Seventh IEEE International Symposium on Multimedia (ISM'05)
pp. 540-544
Investigation of Combining SVM and Decision Tree for Emotion Classification
Thao Nguyen, University Avenue,Lowell, MA
Mingkun Li, DOE Joint Genome Institute, Walnut Creek, CA
Iris Bass, Macomb Community College Warren, MI
Ishwar K. Sethi, Oakland University, Rochester, MI
Full Article Text:
 
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ISM.2005.72
Send link to a friend
| Abstract |
|
This paper discusses the use of a combination of support
vector machine and decision tree learning for recognizing
four emotions in speech, which are Neutral, Angry,
Lombard, and Loud. The base features selected were pitch,
derivative of pitch, energy, speaking rate, formants, bandwidths,
and Mel Frequency Cepstral Coefficients. Three
methods of combining learned support vector machine and
decision tree classifiers were proposed, namely, minimum
misclassification, maximum accuracy, and dominant class.
Using the Speech Under Simulated and Actual Stress database,
the average accuracy from the minimum misclassification,
maximum accuracy, and dominant class methods were
72.4%, 70.8%, 71.3% respectively as opposed to 63.9% and
67.4% which were obtained by using support vector machine
and decision tree learning alone.
|
 Additional Information
|
Citation:
Thao Nguyen, Mingkun Li, Iris Bass, Ishwar K. Sethi,
"Investigation of Combining SVM and Decision Tree for Emotion Classification,"
ism,
pp. 540-544,
Seventh IEEE International Symposium on Multimedia (ISM'05),
2005
|
|
