What is Semantic Search?

Semantic Search, explained

Traditional keyword search looks for documents containing the same words as the query. Semantic search converts both the query and the documents into embeddings, numerical representations of meaning, and returns results that are close in meaning even when the wording differs.

For example, an employee searching an internal knowledge base for how to get reimbursed for travel could find a policy titled Expense Claims Procedure, even though the words reimbursed and travel never appear in its title. Semantic search often uses a vector database to store embeddings and is a core part of retrieval-augmented generation, where relevant passages are retrieved and passed to a language model. Many systems combine semantic and keyword search, because exact terms such as product codes still matter.