Semantic search
A search technique that retrieves results based on meaning and conceptual similarity, computed via embedding vectors, rather than exact keyword matching. The retrieval mechanism underlying RAG, agent memory, and most modern "smart search" features, and the reason personal data ends up embedded and stored in a vector database in the first place.
Defined in 2 GAGE programs, which carry 2 distinct definitions of it. The wording above is taught in AI Data Governance: The Data Chair.
How each discipline defines it
The same term does different work depending on who is using it. These are the definitions as each program teaches them, unedited.
A search method that retrieves conceptually related content even when the exact keywords do not match, as opposed to keyword search, which only finds literal word matches. Semantic search is why an AI legislative tool can find a bill like NYC's Local Law 144 even when a researcher searches for "algorithmic hiring" and the bill's actual title never uses that phrase.
A search technique that retrieves results based on meaning and conceptual similarity, computed via embedding vectors, rather than exact keyword matching. The retrieval mechanism underlying RAG, agent memory, and most modern "smart search" features, and the reason personal data ends up embedded and stored in a vector database in the first place.
Where it is taught
The exact lessons this term appears in. The first 7 topics of every program are free with a free account.
- Embeddings are data too: the vector store as a personal-data system · Feeding the Machines, AI Data Governance: The Data Chair
Terms it appears with
Not an alphabetical neighbourhood: these are the terms taught in the same lessons, ranked by how often they appear together.