TY - GEN
T1 - A learning algorithm for improving the classification speed of support vector machines
AU - Guo, Jun
AU - Takahashi, Norikazu
AU - Nishi, Tetsuo
PY - 2005/12/1
Y1 - 2005/12/1
N2 - A novel method for training support vector machines (SVMs) is proposed to speed up the SVMs in test phase. It has three main steps. First, an SVM is trained on all the training samples, thereby producing a number of support vectors. Second, the support vectors, which contribute less to the shape of the decision surface, are excluded from the training set. Finally, the SVM is re-trained only on the remaining samples. Compared to the initially trained SVM, the efficiency of the finally trained SVM is highly improved, without system degradation.
AB - A novel method for training support vector machines (SVMs) is proposed to speed up the SVMs in test phase. It has three main steps. First, an SVM is trained on all the training samples, thereby producing a number of support vectors. Second, the support vectors, which contribute less to the shape of the decision surface, are excluded from the training set. Finally, the SVM is re-trained only on the remaining samples. Compared to the initially trained SVM, the efficiency of the finally trained SVM is highly improved, without system degradation.
UR - http://www.scopus.com/inward/record.url?scp=33748997961&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33748997961&partnerID=8YFLogxK
U2 - 10.1109/ECCTD.2005.1523140
DO - 10.1109/ECCTD.2005.1523140
M3 - Conference contribution
AN - SCOPUS:33748997961
SN - 0780390660
SN - 9780780390669
T3 - Proceedings of the 2005 European Conference on Circuit Theory and Design
SP - 381
EP - 384
BT - Proceedings of the 2005 European Conference on Circuit Theory and Design
T2 - 2005 European Conference on Circuit Theory and Design
Y2 - 28 August 2005 through 2 September 2005
ER -