1 Introduction
In many classification problems samples from one class are extremely rare (the minority class), while the number of samples belonging to the other class are plenty (the majority class). This situation is known as the rare class problem. It is also referred to as an unbalanced or skewed class distribution problem. Rare class problems naturally arise in several application domains, for example, fraud detection, customer churn, intrusion detection, fault detection, credit default, insurance risk and medical diagnosis.