25 years of The Simpsons, turned into word data
Boston Globe - 09/10/2014
Northeastern historian Ben Schmidt specializes in finding patterns in large bodies of digital text. Usually his work pertains to serious topics, but on a lark—in a single evening, in fact—he pulled closed captioning text to create a database of every line of (nearly) every Simpsons episode from the show’s 25-year run. His search tool, which you can visit online, shows the frequency with which words have appeared over the course of 550 episodes, and also lets you see where in an episode a word occurs , along with the longer dialogue in which it appears.