Glossary
Semantic Search
Searching by meaning using embeddings, often combined with keyword search.
Semantic search finds content by meaning rather than by matching exact words. It compares the embedding of a query against the embeddings of stored passages, so a question phrased differently from the source text still surfaces the right answer. Hybrid search pairs this with traditional keyword matching to catch exact terms that pure semantic search can miss, such as product names, error codes, and SKUs.
CertifChat retrieves with a hybrid approach: a pgvector similarity search for meaning and a Postgres full-text search for precise terms run on every question, and reciprocal rank fusion merges the two rankings into one before the top chunks go to the model. This combination is more robust than either method alone.
See how CertifChat uses it to answer from your documentation.
Related terms
- EmbeddingsNumeric vectors that capture the meaning of text so similar passages sit close together.
- Vector DatabaseA store optimised for finding the embeddings most similar to a query vector.
- Retrieval-Augmented Generation (RAG)Grounding an LLM's answers in retrieved source documents instead of its training data alone.
- ChunkingSplitting documents into passages small enough to embed and retrieve precisely.
Put your knowledge to work today
Connect your sources, drop one snippet on your site, and start deflecting tickets. Start on Starter: 7 days for a one-time $9.