Computer Scientist Says He Can Predict Hits

Oklahoma State University information scientist Ramesh Sharda has reportedly developed a computer system that can tell studios whether a movie will be a box-office hit even before a foot of film has been shot. According to Britain's New Scientist magazine, Sharda's "artificial neural network" analyzes seven key parameters: the "star value" of the cast; the movie's MPAA rating; its release date and competition; its genre; use of special effects; whether it is a sequel; and the number of screens on which it will be shown. Applying that data to 834 movies released between 1998 and 2002, Sharda claimed that the system precisely predicted the films' earnings 37 percent of the time and within a close range 75 percent of the time.
2005-12-20 17:53:40