Weighted SMOTE-Ensemble Algorithms: Evidence from Chinese Imbalance Credit Approval Instances

Abstract

The current study proposes a novel ensemble approach rooted in the weighted synthetic minority over-sampling technique (WSMOTE) algorithm being called WSMOTE-ensemble for skewed loan performance data modeling. The proposed ensemble classifier hybridizes WSMOTE and Bagging with sampling composite mixtures (SCMs) to minimize the class skewed constraints linking to the positive and negative small business instances. It increases the multiplicity of executed algorithms as different sampling composite mixtures are applied to form diverse training sets. Based on the fitted evaluation measures, finally this study recommends that the 'WSMOTE-ensemblek-NN' methodology generating from the WSMOTE-decision tree-bagging with k nearest neighbor is the best fusion sampling strategy which is a novel finding in this domain.