3.3Standard
Measure a customer-support summarizer.
Find quality differences by language, region, product, and input length.
Choose two appropriate practices.
× Incorrect
Measure important slices and inspect sample size, uncertainty, and label quality so average performance does not hide harm.
Detailed explanation
IncorrectDelete slices with poor results
The issue would be hidden.
CorrectMeasure quality and safety separately by important attributes and conditions
Averages can hide subgroup failures.
CorrectCheck sample size, uncertainty, and label quality for each slice
Small or noisy slices need careful interpretation.
IncorrectIgnore slice differences when the global average is high
Some users can still experience serious problems.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 3.3・4.1のグループ別評価とバイアスを確認する。
Official references
What to confirm
全体平均に隠れる条件別の品質差を調査できる。
Key points
- Slices
- Uncertainty
- Average limits
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
Slice analysis
Measuring model behavior for a defined subset such as language, region, or input condition.
Confidence interval
A range expressing uncertainty around an estimated metric.
Question IDAWS-AIF-167
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