Case & CodeA FIELD GUIDE TO AI
Start learning
Intermediate

Semantic search

In plain words

Searching by meaning rather than by exact words, so 'delay' can find 'postponed'.

Words placed on a map by meaning
delivery postponedshipment delayedweather forecastbirthday cake

An embedding turns a piece of text into a position on a map. Texts that mean similar things land close together, even with different words. Searching by meaning means looking for the nearest neighbours on the map.

The fuller definition

Semantic search aims to retrieve material related in meaning to a query, often using embeddings or language models.

Go deeper

Similarity as a map: a deliberately invented two-dimensional example
delivery postponedshipment delayedweather forecast

Related phrasing can be nearby in an embedding space. Real embeddings have many dimensions, and the model determines similarity. Nearby does not mean legally relevant or factually correct.

A concrete example

A search for delivery postponement may find a passage saying shipment was delayed even without the exact wording.

The distinction that matters

Meaning-based matching can miss exact identifiers or return thematically similar but irrelevant material.

Related terms

Where it is taught

How document-grounded AI answers work

Source and further reading

Sentence Transformers · Semantic search

The definition and example are written for this library. The link gives background, official documentation or legal authority as labelled. Legal scope depends on the jurisdiction and current text; product features depend on release and settings.

Lessons, terms, practical guides and sources ESC