In January 2008 I came across a paper from MIT by Antonio Torralba, Rob Fergus, and William T. Freeman. It was about a visualization of all the nouns in the English language arranged by semantic meaning, machine intelligence and computer vision.
They used a total of 7,527,697 images obtained using Google’s Image Search, each tile being the average of 140 images revealing the dominant visual characteristics of each word. For some, the average turns out to be a recognizable image; for others the average is a colored blob. The list of nouns was taken from Wordnet, a database compiled by lexicographers which records the semantic relationship between words.
The mosaic poster they composed is an average of images relating to one of 53,463 nouns. Large-scale groupings correspond to broad categories such as plants or people. Within the plant cluster, for example, tighter semantic groupings are visible such as flowers or trees.