Research On Film Copyright Value Evaluation Based On BP Neural Network Mode

As a mainstay of the entertainment industry,the film industry is an important embodiment of the national cultural soft power.In recent years,there are endless problems in film copyright transaction and infringement compensation,and the evaluation of the value of film copyright in China is shortcomings.More accurate evaluation of the value of film copyright has become an urgent practical problem to be solved in the current evaluation field.The main way to realize the copyright value of China’s film market is different from that of other countries with the United States and other complete film market development.This paper combines the current situation of the domestic film market,It is clear that the main form of the realization of film copyright value in China is the value of the screening right,Then the evaluation of film screening rights: in terms of film box office revenue forecast,Scholars mainly use multiple linear regression and machine learning two kinds methods;In terms of the determination of the sharing rate,Scholars often directly judge or use the industry average based on experience;On the choice of the discount rate,Most scholars choose to use risk accumulation addition to confirm the discount rate;In terms of the earnings period forecast,Scholars often make a brief judgment based on the normal film release cycle or confirm it directly based on the relevant key cycle when the film is released.Based on the characteristics of film copyright value,after understanding the theoretical basis of the film industry,this paper lists the current commonly used methods in the practice of intangible assets evaluation,and analyzes the application degree of various evaluation methods in the evaluation of film copyright value one by one,and finally decides to take the income method as the main evaluation framework.This paper with BP neural network algorithm as the main technical means to predict film box office revenue and square sharing rate,selected 331 non-animated films from 2011 to 2021 as a sample,23 factors to build BP neural network prediction model,and then about film box office revenue and party sharing rate,by random 20 research samples,compare the error rate of different methods to confirm BP neural network model is more accurate.This paper choose risk tired addition as a method to determine the discount rate,the risk-free return rate using one-year Treasury bonds,select investment can cover most film classification of five average yield of listed companies minus risk-free return as risk return,choose scholars generally recognized5% as an individual risk adjustment coefficient.In terms of determining the period of revenue,this paper determines the revenue period of the evaluation object by comparing the screening period of similar released films,and estimates the daily box office revenue curve.Finally,according to the above technical methods,the actual case of The Apostle Walker 2 is studied to predict the above parameters and calculate the evaluation results of its copyright value.The innovation of this paper is the source of the IP factors,further consideration of different IP forms of consumers viewing incentives,also considering in addition to IP can improve the audience familiarity,motivate consumers watching other factors,such as excellent film sequel,historical reality of objective factors,further strengthen the accuracy of the box office revenue forecast.In the study of revenue period,this paper describes the relatively real and accurate daily revenue curve during the release cycle,so as to apply it to the subsequent discount to obtain more accurate evaluation results.Based on the shortcomings of this paper and the difficulties encountered in the process of writing the paper,this paper has the following suggestions for the future research on the evaluation of film copyright value: to further improve the components of film copyright value,strengthen the prediction of the value of film related derivatives,and strengthen the quantification of the individual risk coefficient of each film.