Name two explainability techniques.

• A. SHAP/feature importance
• B. Partial Dependence Plots
• C. LIME/Residual analysis
• D. SHAP/feature importance and Partial Dependence Plots (PDP)

Answer: (D)

Understanding explainability means looking at both why a single prediction was made and how features influence the model overall. SHAP is a local explainability method: it assigns a contribution value to each feature for a specific prediction, so you can see exactly which features pushed that decision up or down. Partial Dependence Plots provide a global view: they show the average effect of a feature on the model’s predictions across many instances, revealing how changes in that feature tend to influence outcomes and highlighting non-linear relationships. Together, SHAP and PDP give a complete picture—per-instance reasoning plus the big-picture behavior of the model. Other options mix a single technique with something unrelated or exclude a local, instance-level explanation, so they don’t offer the same balanced insight.