● SHAP - For Global Feature importance Beewarm plot
● LIME (Local Interpretable Model-Agnostic Explanations) - For Local explanations.
Objective: To predict how prone a person is to get stroke?
Models explored: RF, XGBoost, Logistic regression
XAI-General shap plot: Which features are more important to predict stroke and to predict if stroke is probable or non-probable
XAI LIME subject specific plot: For a specific subject why they are probable or non probable for stroke.
