SHAP
SHapley Additive exPlanations — a method based on game theory that explains individual predictions by calculating each feature's contribution to the prediction. SHAP values are additive and consistent.
Why It Matters
SHAP is the most widely used model explanation technique. It provides both local (per-prediction) and global (overall model) interpretability.
Example
A SHAP analysis showing that for a specific house price prediction of $500K, the pool contributed +$30K, the location +$80K, and the small lot size -$20K.
Think of it like...
Like splitting a restaurant bill fairly — SHAP calculates exactly how much each person (feature) contributed to the total bill (prediction).
Related Terms
Explainable AI
The subfield focused on making AI decision-making processes understandable to humans. XAI techniques provide insights into why a model made a specific prediction.
Interpretability
The degree to which a human can understand the internal mechanisms and reasoning process of a machine learning model. More interpretable models allow deeper inspection of how they work.
LIME
Local Interpretable Model-agnostic Explanations — a technique that explains individual predictions by approximating the complex model locally with a simple, interpretable model.