AI & Technical
Embeddings
Numerical representations of text that allow AI to measure semantic similarity.
Embeddings are dense vector representations of text, where similar meanings cluster together in high-dimensional space. They are the mechanism that allows a RAG system to find relevant knowledge base chunks when a visitor asks a question — even if the exact words don't match. For example, 'how much does it cost?' and 'what are your pricing plans?' produce similar embeddings and retrieve the same pricing documentation. Embeddings are generated by embedding models and stored in vector databases.
Related Terms
Retrieval-Augmented Generation (RAG)
Enhancing AI responses by retrieving relevant documents before generating an answer.
Vector Database
A database optimised for storing and searching embedding vectors at scale.
Knowledge Base
A curated set of documents and FAQs that an AI agent uses to answer questions.