How It Works
Embeddings
Numerical representations that convert text, images, or other data into vectors of numbers, positioned so that similar items end up near each other — the basis for how AI systems compare meaning.
Origin
A long-running idea in computational linguistics; the modern word-embedding era is usually dated to the 2013 word2vec work by Tomas Mikolov and colleagues at Google, not to a coining of the term itself.