Word2vec represents words as vectors whose components encompass what the words mean. Algebraic operations on those vectors, like "king" - "man" + "woman", yield intuitively correct results, like "queen". This project provides the word vectors for the 1000, 5000, 10000, and 25000 most common English words in JSON. github
pages/word2vec-json
http://localhost:3000/assets/pages/word2vec-json/index.html HEIGHT 400
Same demo at http://turbomaze.github.io/word2vecjson/.
pages/word2vec-json/data
pages/word2vec-json/assets/js