SHAP
Definition
SHAP explains model predictions through computationally assigned contributions of individual features.
A feature attribution method that decomposes model predictions via Shapley values from cooperative game theory.
Simply explained
An explanation shows which inputs shifted a prediction up or down.
Why it matters in practice
The term helps to select, develop, evaluate AI systems in a well-founded manner and realistically assess their technical risks.
Example from business
A credit prediction is broken down into contributions of individual input features.
Limits and pitfalls
- Interpreting SHAP values causally, ignoring background distribution, or deriving global statements from a single case.
Origin & history
Lundberg and Lee introduced SHAP in 2017 as a unified approach for additive explanations.
Use in mid-sized companies
- Possible application: A credit prediction is broken down into contributions of individual input features.
Frequently asked questions
What does SHAP mean?
A feature attribution method that decomposes model predictions via Shapley values from cooperative game theory. What mistakes should be avoided? Interpreting SHAP values causally, ignoring background distribution, or deriving global statements from a single case.
Sources
Source checked on 10.10.2026. Editorial application examples are hypothetical; no evidence of real customer results.
Editorially reviewed on: