A technique that converts discrete data like words, images, or categorical variables into dense vector representations in continuous space. Embeddings capture semantic relationships and similarities, enabling machine learning models to process and understand complex data relationships more effectively.
