On the Classification of Imbalanced Datasets

The Classification of Imbalanced Data Sets have received considerable attention in recent research. In this paper, we present an overview of the problem of imbalanced data sets, explain the most commonly used techniques such as sampling and cost sensitive learning, present some evaluation metrics used on imbalanced data sets, quote some interesting points drawn from various popular and latest research papers related to imbalanced classification problem. This paper does not mention all the available research solutions, but try to give a clear picture of imbalanced data set classification problem and present a brief review of existing solutions on this problem. Here, we consider binary classification problem on imbalanced data sets.