Semantic Search
What is semantic search?
Semantic search is a search approach that finds results based on meaning and context rather than exact keyword matching. Using embeddings—vector representations of content—semantic search can identify material that is semantically related to a query even when it doesn't contain the exact same words.
How does semantic search differ from keyword search?
Traditional keyword search (also called lexical search) looks for exact matches between query terms and document content. If you search for "interview techniques" but a document uses the phrase "conversation methods," keyword search might miss it.
Semantic search handles this differently. It allows queries to be more human-like, conversational, and oriented towards natural questions instead of carefully chosen keywords. Instead of matching exact words, semantic search finds content that is conceptually similar to what you're asking about.
For example, with semantic search you can enter a full question like "How do I improve my customer conversations?" and get relevant results about interview skills, even if those exact words don't appear in the content.
How does semantic search work with RAG?
Semantic search often serves as the foundation for retrieval-augmented generation (RAG) systems. When a user asks a question, semantic search retrieves vectorized data from a knowledge base—pulling back content chunks that are semantically related to the query terms.
This retrieved content is then added to the LLM's prompt, enabling it to generate responses informed by relevant information from the knowledge base. The semantic layer makes it possible to find and retrieve meaningful content based on concepts rather than requiring exact keyword matches.
Learn more:
- Turning Disruption into Opportunity: The Stack Overflow AI Story with Ellen Brandenberger
- Inside eSpark's AI Teacher Assistant: RAG, Evals, and Real Classroom Needs
Related terms: