Split procedures into chunks for retrieval.
Keep the meaning of conditions and steps in search results.
Choose two appropriate practices.
Respect semantic boundaries and evaluate chunk size and overlap with representative queries to balance quality and cost.
Detailed explanation
The trade-off between context, recall, and cost can be measured.
Procedures and conditions remain understandable in retrieved chunks.
Retrieval, context size, and cost can suffer.
The values may lose their meaning.
Check official documentation
Compare the following points with the official documentation.
Amazon Bedrock Knowledge Bases公式のデータ取り込み・チャンク化とAIF-C01 Domain 3.2を確認する。
Official references
What to confirm
チャンク境界、重複、サイズのトレードオフを説明できる。
Key points
- Semantic units
- Size
- Overlap
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
Chunk
A bounded fragment of a document stored for embedding and retrieval.
Chunk overlap
Content repeated between adjacent fragments to reduce context loss at a boundary.