In my blog series on my Artificial Intelligence Project, this is post #2.
Well, as is the case with any large project, it is looking like it is going to get very complicated, very quickly. I started flowcharting the program flow and laying out class diagrams and am starting to realize why I can’t already talk to my fridge. However, I am not alone in the quest to talk to appliances. In fact, there are annual competitions where developers try to create the most realistic chatter-bots.
In an article written in the 1950’s, Alan Turing posed the question, “Can machines think?”. In the article, his answer was affirmative, but posed the next question, “If a computer could think, how could we tell?”. To answer that question he designed the Turing Test. In the test, a series of judges chat (through text) to either a computer or another person. If the computer can fool the judges into thinking that they are chatting with another person, then they pass.
In 1990, Hugh Loebner created the Loebner Prize for artificial intelligence. This was based upon the ideas of the Turing Test. He pledged $100,000 (and the gold medal) to the entry that was indistinguishable from a human being. This prize has yet to be won, however, each year prizes are given to the most human-like entry.
I am hoping that by exploring the various approaches that people have used to simulate human interactions will help me to better layout the software that I am writing.
