Non-classical Imbalanced Classification Problems

Abstract

Most of the research in class imbalance are carried out in standard (binary or multi-class) classification problems. However, in recent years, researchers have addressed new classification frameworks beyond standard classification in different aspects. Several variations of class imbalance problem appear within these frameworks. This chapter reviews the problem of class imbalance for a spectrum of these non-classical problems. Throughout this chapter, in Sect. 12.2 some research studies related to class imbalance where only partially labeled data is available (SSL) are reviewed. Then, in Sect. 12.3 the problem of label imbalance in problems where more than a label can be associated to an instance (Multilabel Learning) is discussed. In Sect. 12.4 the problem of class imbalance when labels are associated to bags of instances, rather than individually (Multi-instance Learning), is analyzed. Next, Sect. 12.5 refers to the problem of class imbalance when there exists an ordinal relation among classes (Ordinal Classification). Finally, in Sect. 12.6 some concluding remarks are presented. 12.1 Introduction Over the past several decades, new real-life problems have motivated the development of new classification frameworks that go beyond standard supervised classification [30]. These frameworks have different constraints on the access or nature of supervised data, such as (1) the inherent relationship between instances and labels of a problem, which may be beyond the one-instance one-label standard, (2) the access of partial class information for the training examples, (3) an ordinal relationship among classes. Class imbalance may have different causes and consequences within these frameworks. Therefore, new techniques for dealing with peculiarities and specificities of these problems are necessary. This chapter discuss some research related to imbalanced data when not all class labels of the instances are available (Semi-supervised learning), when instances are associated to more than one label (Multilabel learning), when the label is associated to a bag of instances (Multi-instance learning), when classes are ordered (Ordinal classification) and for regression problems.