The Prediction Model of Weekend Box Office Based on Big Data

The article was based on 56 films released from 2016 to 2017. Firstly, we got the movie related data through official release. Then using crawler software to get weibo data related to movies Secondly combining the movie data and weibo data, we analyzed the correlation between box office and many variables. Using linear regression model, BP neural network and grey model to predict the weekend box office. Finally, we combined the three models to get a combined model to forecast weekend box office. The results show that the relative error of the weekend box office of the combined model is the least, and the prediction method proposed in this paper is effective.