It’s not simple, and there’s a lot of hand waiving involved, but an IBM researcher has published a guide to building your own "Watson Jr." using only commodity hardware and open source software. [Previously 1, 2, 3, 4, and 5] [via]
The Watson Artificial Intelligence system has been discussed on MeFi before. The Jeopardy AI will get a chance to prove its skills in early February when it squares off against Jeopardy titans Ken Jennings and Brad Rutter for the prize of a million dollars.
Alex, I'll take "Kickass Text Analysis Algorithms" for a thousand, please. "Over the rest of the day [at IBM labs] Watson went on a tear, winning four of six games. It displayed remarkable facility with cultural trivia (“This action flick starring Roy Scheider in a high-tech police helicopter was also briefly a TV series” — “What is ‘Blue Thunder’?”), science (“The greyhound originated more than 5,000 years ago in this African country, where it was used to hunt gazelles” — “What is Egypt?”) and sophisticated wordplay (“Classic candy bar that’s a female Supreme Court justice” — “What is Baby Ruth Ginsburg?”)." Next up, a live match up against human winners of Jeopardy. But of course the real question is how good are you? Can you beat Watson?
IBM Research is planning on working on taking artificial intelligence beyond master-level Chess (previously), and on to question answering with a computing system that has been in development for the past two years. Named "Watson," after the I.B.M. founder, Thomas J. Watson Sr., the system will challenge human contestants at Jeopardy (previously). Watson's success depends as much on its ability to understand and respond to the subtleties of human language as it does on the extent of its knowledge database. Don't worry, Alex Trebek knows what's in store. (via)