Influencing Factors And Prediction Of Domestic Films

In recent years,the film industry has developed rapidly,and the annual box office has continued to rise.It has also played an important role in attracting investment,providing employment opportunities,and stimulating consumption growth,bringing social and economic benefits.As an important form of culture,film occupies an important position in the cultural industry and is an important way for cultural exchanges between countries and regions.However,as a major movie country,China has a small number of high-quality movies,and its quality and quantity development are not balanced.Compared with the mature Hollywood movie market in the United States,its cultural output capacity and influence are insufficient.Therefore,studying the influencing factors andpredictions of domestic movie box office will help to enhance the influence of domestic movies in the increasingly fierce market competition,provide reference value for movie investment decisions and reduce distribution risks.This article selects 998 domestic films released in China from the first quarter of 2011 to the first quarter of 2019 as the research object.Based on the current status of the development of the Chinese film market,and existing literature research,the index system of the box office influencing factors is composed of night main factors constructed with a total of 28 indicators.The night main factors are movie genre,audience familiarity,online word of mouth,director influence,actor influence,publicity effect,distribution companies,technical effect,release schedule.The research data comes from Douban.com,Mtime.com,China Film.com,BaiduIndex.com and other official websites.Python crawlers are used to write python programs to collect data.First of all,this paper deals with the data of related indicators.Secondly,by constructing multiple linear regression and random forest regression models,we can obtain variables that have a significant impact on the box office and variable importance measure,and use models to predict the box office revenue of 10 films in order to verify the accuracy of the random forest model and multiple regression model.The results show that the random forest model is more accurate for box office predictions.Then,based on the research results,this paper makes further analysis of the influencing factors of movie box office.Finally,based on the conclusions,reasonable suggestions are put forward,in order to provide decision-making basis for film investment and movie scheduling.The conclusions of this study are as follows:(1)Movie genres include action,comedy,thriller,and fantasy that have a positive impact on the movie box office,,while feature have a negative impact on the box office.Combinations of suspense and thriller have a negative impact on the box office,while combinations of genres like action and comedy,comedy and fantasy have a positive impact on the box office.In movies containing three types,the three types of action,plot,and crime,or the three types of comedy,fantasy,and romance,can both have a positive impact on the box office.(2)Movies based on sequel series or IP adaptations have a positive effect on the box office.(3)The higher the movie’s score,the more it promotes the box office.(4)Actors with high acting skills or high popularity can significantly improve the movie box office.(5)The stronger the director’s ability and the higher thedirector’s popularity,the more the movie box office will promote.(6)Star private companies,Internet companies,and online ticketing platforms can participate in movie distribution,which can effectively increase the box office of the movie.(7)The greater the publicity within a week before the movie’s release,the more it will help the movie box office to increase.Temporary publicity will be effective.(8)Films with formats including Chinese giant screen are conducive to the improvement of the box office.(9)The impact of the release date on the box office of the movie is not significant.Butthe combination of some movie genres and dates can have a positive impact on the box office.For example,the combination of fantasy,comedy,adventure and action films and New Year’s files,and the combination of comedy and action films with summer files have a positive impact on the box office.The main innovations of this article include:(1)Enlarging the sample of domestic research.This article selects 998 films released fromthe first quarter of2011 to the first quarter of 2019 for analysis.Compared with the existing literature with a sample size between 300 and 400,the sample sizehas doubledand the cross-year is longer.The current status of China’s film market is better represented.(2)Richening indicators.In this paper,a total of 28 indicators are selected to build a film box office influencing factor system,including technical effects,distribution companies,and publicity effect.They have not been considered or considered less in previous studies.(3)Extending domestic research methods.The domestic use of random forest classification to predict the box office of movies is more,but the use of random forest regression for numerical prediction is less.This paper adopts the random forest model to provide new research ideas for the predictive analysis of Chinese film box office.