Skip to content

Embeddings

Semantic vectors

The representation of a text as a list of numbers (a vector) that captures meaning, not the exact words. Texts with close meaning have close vectors. It enables semantic search: the assistant finds the relevant fragment even when the user uses different words than the document.

Appeared
Word2Vec — 2013; sentence/document-level embeddings went mainstream from 2022.
Tactics
We turn every fragment of the [knowledge base](#knowledge-base) into embeddings and compare them with the question's embedding to retrieve the most relevant passages.
Related terms

“Embeddings” is part of Glossary · AI Chatbot — part of the Atlas, Websem's reference of AI search, marketing and data terms.