An Empirical Study Of High Box Office Films Based On Data Mining Technology

As an important pillar of the cultural industry,the film industry not only brings huge economic benefits,but also enriches people's leisure life and spiritual world.In recent years,with the development of economy,China's movie market has become more and more prosperous.In 2017,the gross box office revenue in China has reached 55.911 billion yuan,while in many movies,high-end box office movies(with a gross box office revenue of over 50 million)are The box office revenue contribution is the largest.Therefore,it is of great significance for this article to conduct an empirical study of high-priced box-office movies(domestic movies with a box-office receipt of over 50million).It can guide the production of domestic movies.It can tell us what the necessary conditions are for high-priced domestically produced movies.It can give directional guidance to movie makers and lay a foundation for creating more and better domestic movies.This article is based on the research results at home and abroad,according to the actual situation of the domestic film market.Collected data on 434 domestic box office films from 2006 to 2017,and introduced movie types,cooperation with other countries,dates,screen numbers,distribution companies,film formats,adaptations,sequels,directors,screenwriters,actors,word-of-mouth ratings,etc.Looking at the number of these 13 independent variables,the box office receipts of high-priced box-office movies are modeled and analyzed.The results show that the factors that influence the box office domestic movie revenue mainly include: want to see the number of people,film formats,movie types,word-of-mouth ratings,and number of screens.,sequel,distribution company,actor.In this process,the author used a regression support vector machine and a BP neural network algorithm to model the box office receipts of high-priced box office domestic movies to predict the movie box office.It was found that the prediction accuracy rates of the two models were compared.High,but compared with the two,BP neural network prediction accuracy rate is higher.Finally,based on the research results,the author put forward reasonablesuggestions for the development of China-made movies,and provide references for film investment decisions.