Designing a Human Computation Game for Enhancing Early-Phase Movie Box Office Prediction
Movie production is riddled with subjectivity and uncertainty. Each decision made can affect both quality and financial aspects of movies. Previously, various mathematical box office prediction models were proposed, but they focused at the time near the movie release, while earlier predictions would have more benefits to production team. Prediction market was suggested to have good predictability, but it still has some problems. In this study, we designed a human computation game for improving mathematical model performance in early phases which limits what information player knows about the movie at different time and introduces improved mechanics to make the game more similar to the actual movie production. After the experiments, we found that the proposed human computation game did improve mathematical prediction model performance, used in this study, but with limited working conditions. Future work should consider using more complex mathematical models, improving game design, and gathering more data for further validation.
KeywordsBox office prediction Movie industry Business game
- 4.Lash MT, Zhao K (2016) Early prediction of movie success: the who, what, and when of profitability. arXiv preprint arXiv:1506.05382v2Google Scholar
- 5.Mestryán M, Yasseri T, Kertész J (2012) Early prediction of movie box office success based on Wikipedia activity big data. arXiv preprint arXiv:1211.0970Google Scholar
- 7.Panaligan R, Chen A (2013) Quantifying movie magic with Google search. Google, CaliforniaGoogle Scholar
- 8.Doshi L (2010) Using sentiment and social network analyses to predict opening-movie box-office success. Master thesis, Massachusetts Institute of TechnologyGoogle Scholar