Outliers and data descriptions

In previous research the support vector data description (SVDD) is proposed to solve the problem of one-class classification. In one-class classification, one set of data, called the target set, has to be distinguished from the rest of the feature space. In the original optimization of the support vector data description, two parameters have to be given beforehand by the user. In this paper a new, heuristic, error is defined. Minimizing this error, both free parameters in the SVDD can be determined without the use of example outlier objects. This paper shows under what circumstances the heuristic error correlates well with the true error.