Mining Class Outliers: Concepts, Algorithms and Applications

Detection of outliers is important in many applications and has attracted much attention in the data mining research community recently. However, most existing methods are designed for mining outliers from a single dataset without considering the class labels of data objects. In this paper, we consider the class outlier detection problem, i.e., ”given a set of observations with class labels, find those that arouse suspicions, taking into account the class labels.” By generalizing two pioneering contributions in this field, we propose the notion of class outliers and practical solutions by extending existing outlier detection algorithms to detect class outliers. Furthermore, its potential applications in CRM (customer relationship management) are discussed. The experiments on real datasets have shown that our method can find interesting outliers and can be used in practice.