A Survey on Imbalanced Data Learning Method
Qi Zhang · 2005
Many real world applications involve learning from imbalanced sets problems. The class imbalance problem corresponds to domains for which one class is represented by a large number of examples while the other is represented by only a few, such as fraud detection and text classification. We often concentrate on rare cases, which are often of great interest and great value, called the positive cases. Others are called the negative cases. The number of examples of majority class is much higher than the others, which presents an important challenge to the traditional machine learning community. Traditional machine learning algorithms may be biased towards the majority class, thus producing poor predictive accuracy over the minority class. This paper gave several reason of bad performance in rare cases classi- fication and methods for addressing the problems.