MIT EECS - Google Undergraduate Research and Innovation Scholar
Prediction of Future Events with Genesis
Patrick H. Winston
Genesis is an artificially intelligent program that attempts to computationally emulate the way humans read, understand, and analyze stories. It uses Boris Katz’s START system to translate English into a language of relations and events and then uses common sense rules to build a graph of actual and inferred events and causation. Genesis can already compare stories, apply rules, and make inferences about events, characters, and causations, and the next step is to teach it to predict future events given a situation. I plan to do this by analyzing characters’ qualities and what they have done in the past and from there infer their possible next moves. Moreover, I will use story alignment to compare stories that could potentially end the same way.
Through my internships at Originate and MokaFive, I’ve gained a lot of insight on designing and testing systems. Moreover, I’m currently in 6.034, Artificial Intelligence, and am gaining a deep understanding of the mechanics behind my project.”