Li, Yuhua 2011, 'Selecting training points for one-class support vector machines' , Pattern Recognition Letters, 32 (11) , pp. 1517-1522.
![]() |
PDF
Restricted to Repository staff only Download (464kB) | Request a copy |
Abstract
This paper proposes a training points selection method for one-class support vector machines. It exploits the feature of a trained one-class SVM, which uses points only residing on the exterior region of data distribution as support vectors. Thus, the proposed training set reduction method selects the so-called extreme points which sit on the boundary of data distribution, through local geometry and k-nearest neighbours. Experimental results demonstrate that the proposed method can reduce training set considerably, while the obtained model maintains generalization capability to the level of a model trained on the full training set, but uses less support vectors and exhibits faster training speed.
Item Type: | Article |
---|---|
Schools: | Schools > School of Computing, Science and Engineering |
Journal or Publication Title: | Pattern Recognition Letters |
Publisher: | Elsevier B.V. |
Refereed: | Yes |
Funders: | na |
Depositing User: | Yuhua Li |
Date Deposited: | 20 Aug 2015 16:31 |
Last Modified: | 25 Aug 2018 14:54 |
URI: | http://usir.salford.ac.uk/id/eprint/33116 |
Actions (login required)
![]() |
Edit record (repository staff only) |