Classification with imbalanced data : the use of binarization and feature selection

Abstract EN

The imbalanced class problem is related to the real-world application of classification.

It occurs when there is a high difference between the prior probabilities of classes in the learning phase.

It considered as a challenge since it needs to deal with uneven distribution of examples.

The presence of multiple classes implies an additional difficulty since the relations between the classes tend to complicated.

This paper provides a review on multi-class imbalance problem with a focus on feature selection and problem decomposition as a solution of this problem.

Also, it presents a comparative overview of the related solutions that proposed for classification imbalanced datasets.

Moreover, it provides outlines of classification in imbalanced datasets that could help any researcher in this field