Dataset Selection for Training One-Class Support Vector Machines

This paper proposes an efficient training strategy for one-class support vector machines. The strategy 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 neighbors. Experimental results on synthetic and real-world data demonstrate that the proposed training strategy can reduce training set of support vector machines considerably while the obtained model maintains generalization capability to the level of a model trained on the full training set.