基于电影营销因素分析的电影票房预测

本文样本数据选取2011年1月初至2014年12月底公开上映的102部国产电影，选取了8个对票房有影响的营销相关因素作为自变量，票房收入作为因变量，通过建立基于BP神经网络的电影票房预测模型，分别对8个电影营销相关因素进行分析，并对电影的票房进行预测。通过本文，可以看出基于神经网络的票房预测模型有较好的预测效果，对于电影在上映前的投资、导演的选择及营销宣传等等能够提供较好的风险评估，具有较好应用价值。

Modeling on Box-office Revenue Prediction of Movie Based on Factors of Movie Marketing

The date sample of this paper was taken from 102 movies which domestic released from January 2011 to December 2014,the paper selected 8 factors about movie marketing which influenced the film box-office as explanatory variables,and selected film box-office as explained variable. This paper built a model to predict box-office of movies based on neural network and to analyze the influence of 8 factors. According to this paper,we know the model on box-office revenue prediction of movie basing on factors of movie marketing performs better on prediction,so it is good for film production and movie marketing.