UMBC ebiquity

Prediction of Oscar Award Nominations Based on Movie Scripts

Speaker: Niranjan Bhosarekar

Start: Tuesday, July 06, 2010, 10:00AM

End: Tuesday, July 06, 2010, 11:00AM

Location: 325b ITE, UMBC


According to The Numbers, the gross revenue for the Hollywood movie industry was over USD 10 billion in 2009. With annual revenues at this scale, it is critical for a movie to be successful at the box office. High revenue is closely linked to Oscar Award nominations and hence to winning of the award. Building predictive models for Oscar nominations can provide useful insights into predicting Oscar awards. Using a movie script as a movie representation, we retrieve and weight individual movie scenes. We retrieve features from the curve of the movie scene ranking. We introduce a model using Support Vector Machines, which predicts Oscar Award nominations in the Best Screenplay and Best Picture categories based on information from the retrieved features. The model, based on Support Vector Machines, is shown to predict Oscar nominations in the Best Screenplay and Best Picture categories with accuracy greater than 60%.

Committee Members
  • Dr. Charles Nicholas (Chair)
  • Dr. Tim Oates
  • Dr. Tim Finin

Tags: learning

Host: Charles Nicholas