Classification Boosting in Imbalanced Data

Most existing classification approaches assumed underlying training data set to be evenly distributed. However, in the imbalanced classification, the training data set of one majority class could far surpass those of the minority class. This became a problem because classification tends to predict data rewrite by comparing the two classes. This leads to the underestimation of the minority class and influences the performance evaluation criteria. One popular method recently used to rectify this is the SMOTE- Boosting which combines algorithms at data level. Therefore, this paper presents a review of this method by focusing on a two-class problem. Based on the performance criteria of G-mean, the method showed a better performance by taking advantage of the algorithms boost. However, while this affects the performance classification of the base classifier by focusing on all data classes, the SMOTE algorithm alters only for minority classes.