Can Wikipedia Be Used to Predict Future Movie Box Office Revenues?

By David Tobia
November 12, 2012

Researchers have recently used Twitter to forecast election results, changes in stock market prices and box office revenues. Now, researchers are turning to Wikipedia for similar forecasting.
Marton Mestyan of the Budapest University of Technology and Economics in Hungary and his team of researchers have used Wikipedia behavior patterns to predict box office revenues a month before films are released, writes Technology Review.
Number of views, number of human editors, number of edits, and a factor known as collaborative rigor all contribute to the predictions.
“We show that the popularity of a movie could be predicted well in advance by measuring and analyzing the activity level of editors and viewers of the corresponding entry to the movie in Wikipedia,” explains Mestyan.
One problem with the system is that it only seems to work for high-revenue films. Mestyan says the Twitter predictions have similar problems predicting revenues for films that do not do well at the box office.
He says that his method works up to a month in advance, while Twitter only works after the film has been released.
Technology Review cautions that making “predictions” about the past (creating correlations between data sets) is one thing, and actually making accurate predictions about the future is another.

No Comments Yet