MIT researchers developed a technique for cutting through massive amounts of text that combines three popular text-analysis tools — topic modeling, word embeddings, and optimal transport.
— Read on news.mit.edu/2019/finding-good-read-among-billions-of-choices-1220