Credit rating using a hybrid voting ensemble

Abstract

Credit risk analysis is an essential topic in the financial risk management. Credit risk analysis has been the main focus of financial and banking industry. A number of experiments have been conducted using representative supervised learning algorithms, which were trained using two public available credit datasets. The decision of which specific method to choose is a complex problem. Another option instead of choosing only one method is to create a hybrid ensemble of classifiers. © 2012 Springer-Verlag.