Title: Daily box office prediction model based on LSTM
Authors: Yunian Ru - Communication University of China (China) [presenting]
Abstract: The study of movie box offices provides important support for business intelligence decision-making process, such as distribution and cinema management. The task of the daily box office prediction model is to build a dynamic prediction model to rolling forecast daily box office. It is a complex task, as the movie box office has a short life cycle, and the static data and dynamic data that affect the trend of box office are heterogeneous. LSTM recurrent neural network has the ability to memory for a long time, and it can solve the gradient vanishing and exploding problem of RNN. Modeling with LSTM can overcome the shortcoming of ARIMA, it can deal with nonlinear relations, multivariable problems. It can also overcome the shortcoming of traditional ANN models which need to specify the time dependent length. A new model of daily movie box office prediction based on LSTM is proposed. The prediction error MAPE is 30.2\%. The effect of the model is better than that of the previous model. The experiment proved that the more training data collected, the better the prediction effect.