Case study: explaining credit modeling predictions with SHAP

At Fiddler labs, we are all about explaining machine learning models. One recent interesting explanation technology is SHAP (SHapely Additive exPlanations). To learn more about how SHAP works in practice, we applied it to predicting loan defaults in data from Lending Club. We built three models (random, logistic regression, and boosted trees), and looked at several feature explanation diagrams from each. We are not financial experts, so this blog post focuses on comparing algorithms, not insights into loan data. Hopefully, we can bring SHAP into sharper focus.