Binary Information Press
Support Vector Data Description (SVDD) was developed by Tax and Duin to solve one-class classification problems. The method is made robust against outliers in the training set. However, SVDD is particular in the case of one-class classification problem. The authors must construct a classification rule when applied to multi classification. This paper proposes the robust version of SVDD by assigning a weight to each data point, which represents fuzzy membership degree of the cluster computed by the improved PCM method. Accordingly, this paper proposes the multi classification algorithm based on the robust weighted SVDD, and gives the simple classification rule.