Parts | Days | Selection | Search | Updates | Downloads | Help
AGjDPG: Arbeitsgruppe junge DPG
AGjDPG 4: Big Data (joint with SOE)
AGjDPG 4.5: Talk
Thursday, March 14, 2013, 11:15–11:30, H37
Big data; fame and money, box office prediction based on Wikipedia activity data — Márton Mestyán1, •Taha Yasseri1,2,3, and János Kertész1,3,4 — 1Institute of Physics, Budapest University of Technology and Economics, Budapest, Hungary — 2Oxford Internet Institute, University of Oxford — 3Department of Biomedical Engineering and Computational Science, Aalto University, Aalto, Finland — 4Center for Network Science, Central European University, Budapest, Hungary
Use of socially generated Big Data to predict the collective reaction of individuals in societies to a certain event or product has become of great interest in recent years. In this work [1], we investigate the possibility of making precise predictions for the financial success of movies, by monitoring activity and the traffic on Wikipedia articles on the movies. We consider a sample of 312 movies released in the USA market in 2010, and show that, by using a minimalistic linear regression model, one could easily outperform the existing prediction methods. Our model, free of any content analysis, reaches a coefficient of determination of 0.92, one month prior to the movie release.
[1] Márton Mestyán, Taha Yasseri, and János Kertész, Early Prediction of Movie Box Office Success based on Wikipedia Activity Big Data, preprint available at: arXiv:1211.0970.