Answer questions from an internal FAQ.
Evaluate whether retrieved evidence is sufficient.
Choose two appropriate evaluation practices.
Check whether necessary evidence appears near the top and separately assess whether the answer is supported by the retrieved sources.
Detailed explanation
This detects missed relevant chunks.
Fluency does not prove evidence quality.
Retrieval quality and generation quality can be diagnosed separately.
Retrieval performance cannot then be measured.
Check official documentation
Compare the following points with the official documentation.
Amazon Bedrock Knowledge Bases公式の検索評価とAIF-C01 Domain 3.2を確認する。
Official references
What to confirm
検索品質と生成品質を分離して評価できる。
Key points
- Recall
- Precision
- Grounding
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
Recall
The share of relevant documents or fragments that a retrieval system successfully returns.
Precision
The share of returned documents or fragments that are relevant to the query.