In this paper, we propose a new feature selection algorithm for multi-class classification. The proposed algorithm is based on Gaussian mixture models (GMMs) of the features, and it uses the distance between the two least separable classes as a metric for feature selection. The proposed system was tested with a support vector machine (SVM) for multi-class classification of music. Results show that the proposed feature selection scheme is superior to conventional schemes.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Tacksung CHOI, Sunkuk MOON, Young-cheol PARK, Dae-hee YOUN, Seokpil LEE, "A GMM-Based Feature Selection Algorithm for Multi-Class Classification" in IEICE TRANSACTIONS on Information,
vol. E92-D, no. 8, pp. 1584-1587, August 2009, doi: 10.1587/transinf.E92.D.1584.
Abstract: In this paper, we propose a new feature selection algorithm for multi-class classification. The proposed algorithm is based on Gaussian mixture models (GMMs) of the features, and it uses the distance between the two least separable classes as a metric for feature selection. The proposed system was tested with a support vector machine (SVM) for multi-class classification of music. Results show that the proposed feature selection scheme is superior to conventional schemes.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E92.D.1584/_p
Copy
@ARTICLE{e92-d_8_1584,
author={Tacksung CHOI, Sunkuk MOON, Young-cheol PARK, Dae-hee YOUN, Seokpil LEE, },
journal={IEICE TRANSACTIONS on Information},
title={A GMM-Based Feature Selection Algorithm for Multi-Class Classification},
year={2009},
volume={E92-D},
number={8},
pages={1584-1587},
abstract={In this paper, we propose a new feature selection algorithm for multi-class classification. The proposed algorithm is based on Gaussian mixture models (GMMs) of the features, and it uses the distance between the two least separable classes as a metric for feature selection. The proposed system was tested with a support vector machine (SVM) for multi-class classification of music. Results show that the proposed feature selection scheme is superior to conventional schemes.},
keywords={},
doi={10.1587/transinf.E92.D.1584},
ISSN={1745-1361},
month={August},}
Copy
TY - JOUR
TI - A GMM-Based Feature Selection Algorithm for Multi-Class Classification
T2 - IEICE TRANSACTIONS on Information
SP - 1584
EP - 1587
AU - Tacksung CHOI
AU - Sunkuk MOON
AU - Young-cheol PARK
AU - Dae-hee YOUN
AU - Seokpil LEE
PY - 2009
DO - 10.1587/transinf.E92.D.1584
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E92-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2009
AB - In this paper, we propose a new feature selection algorithm for multi-class classification. The proposed algorithm is based on Gaussian mixture models (GMMs) of the features, and it uses the distance between the two least separable classes as a metric for feature selection. The proposed system was tested with a support vector machine (SVM) for multi-class classification of music. Results show that the proposed feature selection scheme is superior to conventional schemes.
ER -