Intelligence · Component · semantic-hotel-search
Semantic hotel retrieval
What it does
Embed natural-language preferences and retrieve hotel candidates from a Qdrant vector collection.
Core pattern
Natural-language preference embeddings + candidate matching
- Input
- Travel preference text
- Output
- Ranked hotel candidates and payloads
Key parameters
| Name | Type | Default | Values |
|---|---|---|---|
| query_text | string | — | — |
| collection | — | — | — |
| result_limit | integer | 20 | 1–1000 |
| embedding_model | — | — | — |
| provider | — | — | — |
Maturity and limits
Actual embedding request and Qdrant search implementation. Static source inspection only; no runtime, provider, deployment or test-pass claim.
No root license found; verify reuse rights and media/font licenses. Provider credentials and customer/card information need protected handling.
Used in pipelines
Not part of a pipeline yet; available as a standalone component.
Industry tags
- Travel & hotelsSource-supported industry
For work involving destinations, room rates and bookings, use natural-language preference embeddings followed by candidate matching.
- Manufacturing & industryTransferable use
For unified lookup of spare-part catalogs and equipment records, use natural-language preference embeddings followed by candidate matching.
Skills pack
Contract scaffold and deterministic mock
Download · 10KBSHA-256 3c42e6a65c689ca4137c0a26f7ced7ef923c3859215ae3ca5c481b940e4f059d
Related components
-
Image and text product similarity search
Shared image-text retrieval + similarity tiers and attribute filters
Working 1 pipeline -
Hotel benefit evidence and comparison
Side-by-side benefit facts + commitment status distinction
Working 1 pipeline -
Multilingual hybrid retrieval and online reindexing
Multi-route retrieval fusion + access filters and index switching
Working not yet in a pipeline