This tool finds related words for a given word. It is based on “unsupervised machine learning”, which means that it was trained WITHOUT any human support, just by “reading” and analyzing a huge collection of English news texts.

I want more information

Unsupervised Learning is a recent approach that can make use of huge amount of data. This tool is based on a collection of English news texts, which contained in total 300 billion words. For each word, a mathematical model was automatically generated that represents, in a sense, the “meaning” of the word. Storing these models uses about 7GB memory on our servers.
Works only for English words. Correct capitalization is important, since “Apple” and “apple” are different words.