Research On The Valuation Of Film Copyright Based On BP Neural Network Model

Before 2020,China ’s film industry has developed rapidly,and the film culture industry has become an important part of China ’s economy.The sudden outbreak of a new coronavirus epidemic has broken the original prosperity and development situation of the film industry.The entire industry has fallen into a downturn.In 2022,the box office of the whole year was less than half of 2019.Many cinemas were on the verge of collapse due to their inability to operate normally.Filmmakers were caught in the dilemma of insufficient filming funds and financing difficulties.In order to help producers obtain external investment,it is necessary to evaluate the value of film copyright reasonably,so that investors can figure out the value of film copyright and the risk of investment,so as to enhance the confidence and enthusiasm of film investment.However,for a long time,China ’s film copyright value system has not been perfected,and a standardized evaluation process has not been formed.The value of film copyright is difficult to determine.In the context of the post-epidemic era,this thesis constructs a BP neural network model to try to provide new ideas for the evaluation of film copyright value.This thesis analyzes the source of film copyright value,combined with the reality,the film box office income as an approximate film copyright income.When choosing the evaluation method,by comparing the three basic evaluation methods,it is found that the value evaluation of film copyright is more applicable to the income method,but the prediction of the future income of film copyright is different from the value evaluation of enterprises and physical assets.There is no historical financial data that can be used,which is a major difficulty in the evaluation.In order to reasonably predict the expected future income of film copyright,this thesis constructs a BP neural network model.Combined with the references,according to the classification of internal and external influencing factors of film copyright,the familiarity,plot,comedy,love,action,director,screenwriter,actor,3D,IMAX,schedule,distribution company,word-ofmouth,screening days,epidemic impact and average ticket price are selected as explanatory variables,and the explanatory variables are quantified and normalized.The BP neural network model was used to train the data of 130 films since 2016,and 10 %of the samples were randomly selected to test the multiple linear regression model,which verified the accuracy of the BP neural network model in predicting the future expected income of the film.Finally,the constructed BP neural network model is applied to the evaluation practice to predict the future expected income of the film Home Coming and reasonably determine the sharing rate,discount rate and income period.Combined with the formula of the income method,the copyright value of the film Home Coming is 530 million yuan.Compared with the film copyright value of542 million yuan in multiple linear regression,the error rate of the BP neural network model is smaller and closer to the actual value.