Many of you may remember that IBM’s Deep Blue supercomputer managed to beat Garry Kasparov in a chess match in 1997. (Kasparov disputed the fairness of the match, and played the program to a draw in a later match.)
The New York Times recently carried a report that IBM is planning to take on a new “machine thinking” challenge: putting together software that can compete on the TV game show, Jeopardy!. As a long-time fan of the show, I am most interested to see how this works out. Of course the program has to have a large database of facts available, but it also has to be able to interpret the clues:
The way to deal with such problems, Dr. Ferrucci said, is to improve the program’s ability to understand the way “Jeopardy!” clues are offered. The complexity of the challenge is underscored by the subtlety involved in capturing the exact meaning of a spoken sentence. For example, the sentence “I never said she stole my money” can have seven different meanings depending on which word is stressed.
The planned test of the program, called Watson (after Thomas, not Dr. John H.), will have some elements similar to a Turing test:
Under the rules of the match that the company has negotiated with the “Jeopardy!” producers, the computer will not have to emulate all human qualities. It will receive questions as electronic text. The human contestants will both see the text of each question and hear it spoken by the show’s host, Alex Trebek.
According to the article, they are thinking of inviting Ken Jennings to participate. It should be interesting.