4.1Standard
Review a hiring-support model by demographic group.
Overall averages may hide harm.
Choose two appropriate evaluations.
× Incorrect
Compare error metrics by subgroup, investigate data gaps and proxy variables, and reevaluate after threshold or policy changes.
Detailed explanation
CorrectCompare precision, recall, and error rates for each subgroup
Differences hidden by an overall average become visible.
IncorrectRemove protected attributes and assume bias is gone
Proxy variables and sampling bias may remain.
CorrectCheck missing data, proxy variables, and impact and reevaluate threshold changes
Measurement bias and operational impact can be tracked.
IncorrectDeclare fairness from a high overall accuracy
Serious subgroup differences can be hidden.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 4.1の公平性、バイアス、評価指標を確認する。
Official references
What to confirm
全体指標だけでなくグループ別の差と業務影響を評価できる。
Key points
- Subgroups
- False positives
- Proxy variables
Notes
- Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Subgroup evaluation
Measuring performance and errors separately for defined populations or operating conditions.
False positive
A negative case that a model incorrectly labels or accepts as positive.
Question IDAWS-AIF-205
EnvironmentAWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
Last technical review2026-08-20
Report an error or rights issue