Vector Embeddings
Vector embeddings are numerical representations of words, sentences or entire passages of text in which similar meanings sit close together in mathematical space – the technical basis of semantic search and of many AI applications.
In practice
Instead of searching for exact keywords, systems using embeddings compare the similarity in meaning of whole queries and text passages. That explains why modern search systems and AI answer engines still find relevant content when a question is phrased differently but means much the same thing.