A holistic approach to interpretability in financial lending: Models, visualizations, and summary-explanations

Chaofan Chen, Kangcheng Lin, Cynthia Rudin, Yaron Shaposhnik, Sijia Wang, Tong Wang

Abstract: Highlights•Introduction of a DSS for financial lending that includes several AI approaches that form a holistic approach. A novel globally interpretable two-layer additive risk model, which lends naturally to sparsity, decomposability, visualization, case-based reasoning, feature importance, and monotonicity constraints.•An interactive visualization tool for the model and its local explanations.•An application of our approach to finance, indicating that black-box models may not be necessary in the case of credit-risk assessment.