Retail Search

v0-cranfield-opensearch-baseline

 Live Cranfield baseline

Explain

Inspect how a query becomes an OpenSearch request, how BM25 ranks documents, and what the system returns as explanation evidence.

JSON
  1. 1 Query User submits a short technical search phrase.
  2. 2 Normalize Whitespace is trimmed and repeated spaces are collapsed.
  3. 3 OpenSearch Query A multi_match request targets title, abstract, and text.
  4. 4 Retrieve OpenSearch searches the live cranfield-v0 index.
  5. 5 Rank BM25 scores documents with title and abstract boosts.
  6. 6 Explain The response includes the generated query, decisions, latency, and top-result rationale.

Run explain

Run explain to see the generated OpenSearch query.