Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/28375
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | Dual v-support vector machine with error rate and training size biasing |
Author: | Chew, H. Bogner, R. Lim, C. |
Citation: | 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing : proceedings, 7-11 May, 2001, Salt Palace Convention Center, Salt Lake City, Utah, USA : vol. 2, pp. 1269-1272 |
Publisher: | IEEE SIGNAL PROCESSING SOCIETY |
Publisher Place: | CD-ROM |
Issue Date: | 2001 |
ISBN: | 0780370430 |
Conference Name: | IEEE International Conference on Acoustics, Speech and Signal Processing (2001 : Salt Lake City, Utah) |
Editor: | John Matthews, V. |
Statement of Responsibility: | Hong-Gunn Chew ; Bogner, R.E. ; Cheng-Chew Lim |
Abstract: | Support vector machines (SVMs) have been successfully applied to classification problems. The difficulty in selecting the most effective error penalty has been partly resolved with ν-SVM. However, the use of uneven training class sizes, which occurs frequently with target detection problems, results in machines with biases towards the class with the larger training set. We propose an extended ν-SVM to counter the effects of the unbalanced training class sizes. The resulting dual ν-SVM provides the facility to counter these effects, as well as to adjust the error penalties of each class separately. The parameter ν of each class provides a lower bound to the fraction of support vector of that class, and the upper bound to the fraction of bounded support vector of that class. These bounds allow the control on the error rates allowed for each class, and enable the training of machines with specific error rate requirements. |
Description: | © Copyright 2001 IEEE |
DOI: | 10.1109/ICASSP.2001.941156 |
Appears in Collections: | Aurora harvest 6 Electrical and Electronic Engineering publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.