From Story Line to Box Office: A New Approach for Green-Lighting Movie Scripts
提出一种结合编剧领域知识、自然语言处理和统计学习的方法，仅基于剧本文本预测电影的投资回报率，帮助制片厂更准确地筛选剧本，提升整体盈利。
Movie studios often have to choose among thousands of scripts to decide which ones to turn into movies. Despite the huge amount of money at stake, this process—known as green-lighting in the movie industry—is largely a guesswork based on experts’ experience and intuitions. In this paper, we propose a new approach to help studios evaluate scripts that will then lead to more profitable green-lighting decisions. Our approach combines screenwriting domain knowledge, natural-language processing techniques, and statistical learning methods to forecast a movie’s return on investment (ROI) based only on textual information available in movie scripts. We test our model in a holdout decision task to show that our model is able to significantly improve a studio’s gross ROI.