Disparate Impact

nounid 4853·updated May 18, 2026

verified

For Predictor Y and Sensitive Impact S. Definition 6.2 Disparate Impact (DI) = P[Yˆ = 1 | S != 1]/P[Yˆ = 1 | S = 1]

MWE

Classifications

per-axis with provenance

Metric100%manual revieweval-seeds-v0.3-2026-05-14

Regulated75%llm-generatedllm:claude-haiku-4-5

Pii68%llm-generatedllm:claude-haiku-4-5

3 surface forms

1 sense across 1 framework

A Comparative Study of Fairness-Enhancing Interventions in Machine Learning1 senseview framework →

§1: For Predictor Y and Sensitive Impact S. Definition 6.2 Disparate Impact (DI) = P[Yˆ = 1 | S != 1]/P[Yˆ = 1 | S = 1]

term as subject — 0 triples

No outgoing triples

This term is not the subject of any RDF-style relationship yet.

term as object — 0 triples

No incoming triples

No other term currently asserts a relationship to this one.