Embedding
A numerical representation of text that captures its meaning for comparison and search, enabling AI models to understand semantic relationships.
Definition
An embedding is a vector (list of numbers) that represents text in a way that captures semantic meaning. Similar concepts have similar embeddings, enabling computers to understand that "dog" and "puppy" are related even though they share no letters.
Embeddings power semantic search, recommendation systems, and retrieval-augmented generation by allowing comparison of meaning rather than exact text matches.
Why It Matters
Embeddings enable AI to understand what you mean, not just what you type. This powers search that finds relevant results even when queries don't match exact keywords.
Understanding embeddings helps design better knowledge bases and search systems that surface truly relevant information.
Examples in Practice
A company embeds their entire documentation library, allowing customer support to find relevant articles even when user questions don't use official terminology.
Product embeddings power a recommendation system that suggests items similar to what customers have viewed or purchased.
A legal research tool uses embeddings to find relevant case law based on the meaning of a legal argument, not keyword matching.