Research on Movie Box Office Forecasting Based on Internet Data

The film industry is one of the fastest growing industries in the world, especially the growth of Chinese movie box office. Now the film industry is still a high risk industry. As a capital intensive industry, movie box office forecasting is particularly important to investment and financing of the film, cinema advertising sales, row of film theaters, marketing, operations and various aspects. Based on the traditional data of the film, this paper adds the new media data of Internet to predict the movie box office. Our model involves 16 variables integrated as 5 aspects - film´s transmission capacity, publicity of the film, film marketing efforts, basic attributes of the film, film creators. With the application of BP neural network, the forecast accuracy can reach 72% one month before film release.