The Application Of Machine Learning In The Prediction Of Movie Box Office

With the improvement of social living standards,people pay more and more attention to the pursuit of the spiritual level,entertainment consumption in life has become increasingly large proportion.As an important part of mass entertainment consumption,film not only enriches people's spare time and spiritual world,but also is an important medium of cultural exchange between countries.For a film,the box office is an important measure of the success of the film,investors often need to consider the risks and returns to decide whether to invest in the film.If we can make a reasonable and accurate forecast of the box office,we can reduce the loss of investment risk to a great extent,and adjust the strategy reasonably in the filming,production,promotion and release of the film,so as to maximize the investment income.This paper selects 526 films released in mainland China from 2016-2019 as the research object,and constructs a box-office revenue forecasting index system composed of 8 variables: film type,popular theme,actor,director,film format,release date,film duration and distribution company.At the same time,through comprehensive comparison,we can get four important factors that affect the box office: film length,film format,actors and release time.Finally,according to the results of empirical analysis,some feasible suggestions are put forward for filmmakers,distributors and investors,and the shortcomings of this study are summarized.