Evaluation Of Movie Box Office Based On Random Forest

With the continuous development of China’s film market,the film as a cultural and creative products,the value of the film box office evaluation conforms to the development of the film market.The evaluation of movie box office value is of great significance to reduce the risk of film market investment and promote the healthy development of film industry.At present,the film box office value evaluation system has not yet been fully established in China,so this paper discusses the value of movie box office based on the present situation of film development in China to provide some ideas for the evaluation of movie box office value.In our country,the realization of movie box office value mainly depends on the movie box office income,the evaluation of movie box office value is obtained by calculating the movie box office income,so the analysis and calculation of film box office income is the key point of movie box office value evaluation.In this paper,the value model of movie box office based on random forest is proposed in view of the large input data dimension,the fast convergence speed and the high requirement of prediction accuracy in the movie box office value model.The model introduces the stochastic forest algorithm to improve the information processing capability of the model,and solves the problem that the accuracy of box office valuation is not high due to the variable types of influencing factors of movie box office.A total of 324 films were selected as the dataset.Firstly,based on the index system of domestic and foreign traditional movie box office value model,according to the influence of the internet,the Sina index,search engine index and network media attention are added to the index system.Secondly,the relevant parameters in the random forest model are adjusted to make it more suitable for the box office value model.Finally,the value of the box-office value model of the random forest movie will be collected.The original data is divided into training set and test set,the training and analysis of the model,from the experimental results found in the method based on the box office value model compared to the regression analysis model has a higher accuracy and stability,to meet the box office value assessment needs.