Evaluating subgroup fairness | AWS Certified AI Practitioner (AIF-C01) original practice question

4.1Standard

Review a hiring-support model by demographic group.

Overall averages may hide harm.

Choose two appropriate evaluations.

Hint 1: Focus on the requirement

Identify what must change and what must remain unchanged. Then choose the command whose scope matches that requirement.

Hint 2: Check each command's role

Separate commands that display information, change state, remove data, or affect a child process.

属性グループ別に適合率、再現率、誤検知などを比較する: Differences hidden by an overall average become visible.
データ不足、代理変数、業務影響を確認し、閾値変更後も再評価する: Measurement bias and operational impact can be tracked.
全体の平均精度が高ければ公平と判断する: Serious subgroup differences can be hidden.
保護属性を削除すれば偏りは必ず消える: Proxy variables and sampling bias may remain.

Detailed explanation

CorrectCompare precision, recall, and error rates for each subgroup

Differences hidden by an overall average become visible.

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.

IncorrectRemove protected attributes and assume bias is gone

Proxy variables and sampling bias may remain.

An example you can run in a temporary verification environment.

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。

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。

Foundation review

Check whether the command acts on the current shell, a new process, an existing process, or a file.

Use the supplied command and compare the output with the expected result.

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