Northwestern study: Data analysis trumps critics when picking movies
January 23, 2015
A new Northwestern study shows the best predictor of a movie’s significance is how often it is referenced in other movies, not what critics say about it.
Luis Amaral, leader of the study and co-director of the Northwestern Institute on Complex Systems, systematically compared different approaches for estimating a film’s significance. His team found that its automated method of movie citations was better at predicting greatness than the movie critics.
“Movie critics can be overconfident in spotting important works, and they have bias,” Amaral said in a news release. “Our method is as objective as it gets.”
In the study, Amaral and his colleagues looked at the number of times 15,425 U.S.-produced films appeared in the National Film Registry of the U.S. Library of Congress. They found that for movies 25 years or older, the number of times they were referenced in other movies did the best job at predicting inclusion in the Film Registry.
“Directors keep coming back to movies that are significant,” Amaral said. “If you show a little bit from ‘Psycho,’ such as referencing the shower scene, you are putting that whole movie in front of the viewer of the new movie.”
Researchers also reported that this method could also be applied to scientific papers, music and paintings.
“It can be difficult to distinguish a good scientific paper from an average one, much like the movies,” Amaral said. “My next goal is to develop a good measure of scientific citations to get inside what is going on in the scientific literature.”
Email: [email protected]