A Classification Approach of Neural Networks for Credit Card Default Detection

Abstract

By using a neural network system, which is more complex and sophisticated than a simple linear regression model, the classification simulation shall have a better performance. The data was extracted from UCI machine learning Lab which represents Taiwan credit card defaults in 2005 and their previous payment histories. This article mainly tried to determine the factors that strongly predict the future default probability with neural network comparing with linear model shows the advantage of deep learning in financial area.