A Review on Imbalanced Learning Methods

Nowadays learning from imbalanced data sets are a relatively a very critical task for many data mining applications such as fraud detection, anomaly detection, medical diagnosis, information retrieval systems. The imbalanced learning problem is nothing but unequal distribution of data between the classes where one class contains more and more samples while another contains very little. Because of imbalance learning problems, it becomes hard for the classifier to learn the minority class samples. The Aim of this paper is to review on various techniques which are used for resolving imbalanced learning problem. This paper proposes a taxonomy for various methods used forhandling the class imbalance problem where each method can be categorized depending on the techniques it uses. To handle imbalanced learning problem significant work