Vector (Embedding)
In one sentence
Converting a block of text or an image into a series of numbers. Content with similar meanings gets similar numbers, making it possible to retrieve information by meaning.
Keyword search only matches exact words. Once text is converted into vectors, queries phrased differently but sharing the same intent—like 'how to return an item' and 'can I exchange something bought by mistake'—can also be matched.
Documents in the knowledge base are first split into chunks. Each chunk is converted into a vector and stored in a vector database. When a user asks a question, the system retrieves the most closely matching chunks based on vector similarity.
Example
A customer asks, 'Can you issue a dedicated VAT invoice?' The system uses vector search to find the matching section in the documentation: 'Invoice type: VAT special invoices supported.'
Related terms
- Vector Database
- A database that stores vectors and quickly finds semantically similar content, commonly used as a core component for RAG and knowledge base retrieval.
- RAG (Retrieval-Augmented Generation)
- A method where AI retrieves relevant information from a knowledge base first, then generates an answer based on those materials.
- Text Chunking
- Splitting long documents into appropriately sized chunks before saving them to a knowledge base, making it easy to retrieve only the relevant sections.
