A scientist believes he has invented a computer program that would predict whether a film will be a success at the box office – even before it’s made.

“We are trying to forecast the success of a movie based on things that are decided before a movie has been made,” Professor Ramesh Sharda of Oklahoma State University told Reuters. “Our goal is to try to find oil in a way.”

Sharda, who specializes in information systems, has worked on the model for seven years and analyzed more than 800 films. He applies seven criteria to each movie: its rating, competition, cast, genre, special effects, whether it is a sequel and the number of theatres it opens in. The films are ranked in nine categories, from “flop” (less than $1 million at the box office) to “blockbuster” (more than $200 million).

The neural network’s results were accurate within one category 75 per cent of the time.

Sharda’s research will appear in “Expert Systems With Applications” journal next year. He may also further develop his system with an unnamed Hollywood studio.






Related Stories: