The classification time required by conventional multi-class SVMs greatly increases as the number of pattern classes increases. This is due to the fact that the needed set of binary class SVMs gets quite large. In this paper, we propose a method to reduce the number of classes by using nearest neighbor rule (NNR) in the principle component analysis and linear discriminant analysis (PCA+LDA) feature subspace. The proposed method reduces the number of face classes by selecting a few classes closest to the test data projected in the PCA+LDA feature subspace. Results of experiment show that our proposed method has a lower error rate than nearest neighbor classification (NNC) method. Though our error rate is comparable to the conventional multi-class SVMs, the classification process of our method is much faster.
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
Sung-Wook PARK, Jong-Wook PARK, "Using Nearest Neighbor Rule to Improve Performance of Multi-Class SVMs for Face Recognition" in IEICE TRANSACTIONS on Communications,
vol. E87-B, no. 4, pp. 1053-1057, April 2004, doi: .
Abstract: The classification time required by conventional multi-class SVMs greatly increases as the number of pattern classes increases. This is due to the fact that the needed set of binary class SVMs gets quite large. In this paper, we propose a method to reduce the number of classes by using nearest neighbor rule (NNR) in the principle component analysis and linear discriminant analysis (PCA+LDA) feature subspace. The proposed method reduces the number of face classes by selecting a few classes closest to the test data projected in the PCA+LDA feature subspace. Results of experiment show that our proposed method has a lower error rate than nearest neighbor classification (NNC) method. Though our error rate is comparable to the conventional multi-class SVMs, the classification process of our method is much faster.
URL: https://global.ieice.org/en_transactions/communications/10.1587/e87-b_4_1053/_p
Copy
@ARTICLE{e87-b_4_1053,
author={Sung-Wook PARK, Jong-Wook PARK, },
journal={IEICE TRANSACTIONS on Communications},
title={Using Nearest Neighbor Rule to Improve Performance of Multi-Class SVMs for Face Recognition},
year={2004},
volume={E87-B},
number={4},
pages={1053-1057},
abstract={The classification time required by conventional multi-class SVMs greatly increases as the number of pattern classes increases. This is due to the fact that the needed set of binary class SVMs gets quite large. In this paper, we propose a method to reduce the number of classes by using nearest neighbor rule (NNR) in the principle component analysis and linear discriminant analysis (PCA+LDA) feature subspace. The proposed method reduces the number of face classes by selecting a few classes closest to the test data projected in the PCA+LDA feature subspace. Results of experiment show that our proposed method has a lower error rate than nearest neighbor classification (NNC) method. Though our error rate is comparable to the conventional multi-class SVMs, the classification process of our method is much faster.},
keywords={},
doi={},
ISSN={},
month={April},}
Copy
TY - JOUR
TI - Using Nearest Neighbor Rule to Improve Performance of Multi-Class SVMs for Face Recognition
T2 - IEICE TRANSACTIONS on Communications
SP - 1053
EP - 1057
AU - Sung-Wook PARK
AU - Jong-Wook PARK
PY - 2004
DO -
JO - IEICE TRANSACTIONS on Communications
SN -
VL - E87-B
IS - 4
JA - IEICE TRANSACTIONS on Communications
Y1 - April 2004
AB - The classification time required by conventional multi-class SVMs greatly increases as the number of pattern classes increases. This is due to the fact that the needed set of binary class SVMs gets quite large. In this paper, we propose a method to reduce the number of classes by using nearest neighbor rule (NNR) in the principle component analysis and linear discriminant analysis (PCA+LDA) feature subspace. The proposed method reduces the number of face classes by selecting a few classes closest to the test data projected in the PCA+LDA feature subspace. Results of experiment show that our proposed method has a lower error rate than nearest neighbor classification (NNC) method. Though our error rate is comparable to the conventional multi-class SVMs, the classification process of our method is much faster.
ER -