A Hybrid Credit Scoring Model Using Neural Networks and Logistic Regression
Credit scoring is one of important issues in banking to control a loss due to debtors who fail to meet their credit payment. Hence, the banks aim to develop their credit scoring model for accurately detecting their bad borrowers. In this study, we propose a hybrid credit scoring model using deep neural networks and logistic regression to improve its predictive accuracy. Our proposed hybrid credit scoring model consists of two phases. In the first phase, we train several neural network models and in the second phase, those models are merged by logistic regression. In experimental part, our model outperformed baseline models on over three benchmark datasets in terms of H-measure, area under the curve (AUC) and accuracy.
KeywordsDeep learning Logistic regression Credit scoring
This research was supported by the Private Intelligence Information Service Expansion (No. C0511-18-1001) funded by the NIPA (National IT Industry Promotion Agency) and by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (No.2017R1A2B4010826).
- 11.Van Gestel, T., et al.: Linear and nonlinear credit scoring by combining logistic regression and support vector machines. J. Credit Risk 1(4) (2005)Google Scholar
- 12.Vellido, A., Martín-Guerrero, J.D., Lisboa, P.J.G.: Making machine learning models interpretable. ESANN 12 (2012)Google Scholar
- 15.Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)
- 16.Ruder, S.: An overview of gradient descent optimization algorithms. arXiv preprint arXiv:1609.04747 (2016)
- 17.Asuncion, A., Newman, D.: UCI machine learning repository (2007)Google Scholar
- 18.FICO, Xml challenge, https://community.fico.com/s/explainable-machine-learning-challenge. Accessed 1 Oct 2018
- 19.Arnold, T.B.: KerasR: R interface to the Keras deep learning library. J. Open Source Softw. 2 (2017)Google Scholar