Research About The Influence Of IP Factor On Early Movie Box Office Prediction

In recent years,the domestic and foreign scholars have relatively more research on movie box office.In the industry,according to the nature of the selected variables,the prediction of movie box office can be divided into two categories:early prediction and real-time prediction.The data of early prediction come from the movie itself,and data of real-time prediction has nothing to do with the movie.Early predictions can make a general assessment of the movie box office before filming,and can avoid investment risks and maximize the benefits.This paper mainly studies the influence factors of early movie box office,and predicts the early movie box office.By referring to a large number of research of domestic and foreign scholars,and combining the specific characteristics of the Chinese film box office market,it is found that with the rise of(Intellectual Property,IP)film heat,the film adapted from high quality IP has become a new force to pull the box office of the film.Therefore,domestic scholars have verified about IP is an impact factor that affects the early movie box office.The results show that there is a positive correlation between IP and the film box office.However,the processing of IP adaption variable is rather simple,only the binary representative is adapted from IP.Because the processing method of the IP adaptation variable is difficult,how to quantify the IP factor is a key task.Therefore,this paper proposes an improved Bayesian voting mechanism to quantify the impact of IP adaptation.Here it is called IP quality,and the IP quality is as a factor that affect the early film box office.Finally,three representative models of movie box office prediction are selected,which are Multiple Linear Regression(MLR),support vector machine regression(Support Vector Regression,SVR)and BP(Back Propagation,BP)neural network model respectively.And the three models are used to verify the new IP quality factors.In the indicators that measure the regression results of the model,Coefficient of Determination R~2(Coefficient of Determination,R~2)is used as a regression evaluation index.The experiment results show that,the R~2 of the three models have a corresponding increase with the IP quality factor.Therefore,the IP quality impact factor is positively related to the early movie box office.By increasing the IP quality impact factor,the prediction accuracy of the movie box office can be improved.In the future film box office prediction,we can try to quantify the other factors reasonably,so as to further improve the early film box office prediction,and avoid the investment risk and maximize the benefit.We should invest more in movies that audiences like,so as to make our film market better and better.