Finding Counterfactual Explanations and Local Feature Importance for a Selected Prediction of a Credit Scoring Model

In the example, the Credit Scoring data set is partitioned to training and test samples. Then, the black box model (Neural Network) is trained on the standardly pre-processed training data using the AutoML component. The Workflow Object capturing the pre-processing and the model is provided as one of the inputs for the Local Explanation View component.

The Local Explanation View component is then used to inspect the local model behavior via finding Counterfactual Explanations for a selected single loan applicant and inspecting Local Feature Importance with the Local Surrogate Generalized Linear Model relevant for this selected applicant.

External resources

• Give Me Some Credit - Kaggle data set
• Molnar, Christoph. "Interpretable machine learning. A Guide for Making Black Box Models Explainable", 2019.
• KNIME Integrated Deployment - KNIME.com