Canon term · Stack categories

Vector search

Retrieval by similarity in an embedding space, returning items whose vector representation is closest to a query's.

  • StatusIndustry term
  • AdoptionWidely used
  • LayerIntelligence

Why it matters

Answers 'what looks like this question.' Finance more often needs 'what is connected to this entity, and how' — an exposure chain, a constituent history, a data lineage — which is a traversal, not a similarity search. No amount of embedding quality converts one into the other. A further governance problem: reconstructing a past answer requires the index as it stood at the time, which most deployments cannot produce.

Where it comes from

Approximate nearest-neighbor literature; productized from 2021.

No published instrument defines this term yet; the note above is its origin.

What it is not

These are routinely confused with Vector search. The distinctions have consequences for how a system is governed.

  • Graph traversal

    Similarity finds things that resemble each other. Traversal follows declared relationships. Different questions, different answers.

How it connects

Not to be confused with

Cite and reuse

Cite the identifier QIS-TERM-00045, not the URL: identifiers are stable, URLs may change. This entry is free to read, quote and index under the dual license. Corrections sent to editor@qisfund.com are published.

Technical details: identifier, graph relationships and record
QIS-TERM-00045: outbound relationships as published in api/graph.json and api/lexicon.json
PredicateTarget identifierTargetPredicate meaning
notEquivalentTocat:graph-databaseGraph databaseThe two are routinely conflated and are distinct.
relatedToconcept:graphragGraphragAn association too weak or too general for a stronger verb.
Canonical identifier
QIS-TERM-00045
Tier
Canonical industry term
Domain · layer
Stack categories · Intelligence
Semantic aliases
None recorded
First published
2026-08-02
Last reviewed
2026-08-02 · 180-day cycle