Using Wikipedia to Predict Box-Office Success

August 23, 2013 | Andy Cush

It stands to reason that the more popular a movie’s Wikipedia page is–that is, the more pageviews it gets, the more people who work on editing it, that sort of thing–the more popular the film itself is. While its true that the Wikipedia crowd probably isn’t a perfect microcosm of a movie-viewing audience, there’s likely enough overlap that we can assume that if there’s enthusiasm about a film on Wikipedia, there’s probably enthusiasm about it IRL. But does that hypothesis hold up predictively? Could you use Wikipedia to guess with some accuracy how successful a movie will be, before it even comes out?

According to a group of researchers out of Oxford, the Central European University at Budapest, and Budapest University of Technology and Economics, the short answer is “yes.” The group charted the popularity of a film’s Wikipedia entry one month before its premiere with that film’s box office numbers opening weekend, and eventually came to a mathematical model that could predict financial success based on Wikipedia popularity with 77% accuracy. According to Phys.org, existing models top out at around 57%.

János Kertész of the Central European University of Budapest explained how Wikipedia could be a powerful predictive tool, especially when compared to other social networks. “We have demonstrated for the first time that Wikipedia edit statistics provide us with another tool to predict social events,” he said. “We studied the problem of predicting the financial success of movies and concluded that, in some aspects, forecasting based on Wikipedia outperforms tweets as Wikipedia activity has a longer timescale which enables earlier predictions.”