Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
536515 | Pattern Recognition Letters | 2011 | 6 Pages |
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.
► A novel training points selection method based on local geometry and kNN is developed. ► It selects extreme points which sit on the boundary of data distribution. ► It increases training efficiency and maintains generalisation capability of 1-class SVM.