Research On Film Marketing Strategy Based On Data Mining

The high-speed popularity of the Internet has caused rapid growth in data.Processing data information and extracting the meaning behind the data have become the focus of the community.The development of big data is continuously updating,which promotes profound changes in work and production patterns in different industries.In the film industry,the application of big data technology has already achieved initial results.Using data mining to explore relevant information in the field of film marketing can not only judge the development status,but also help adjust marketing strategies and improve marketing efficiency.Reasonable film marketing strategies can achieve the goal of high box office;therefore,it is necessary to comprehensively consider various factors in the movie movie-making production,the movie release and the movie projection.Using different data mining methods to analyze these factors and explore the reasons can enhance the innovation space of the movie marketing model.For the purpose of propose corresponding marketing strategies,this paper adopts a combination of theoretical analysis and empirical research,the investigation of the film marketing is carried out from audiences and films under the guidance of marketing concepts,and then establishing different models.The main content is as follows:1.Based on the theory of marketing and behavioral economics,the audience behavior model is divided into three categories,including audience decision-making behavior,audience purchase behavior and audience post-purchase behavior,and establishing a corresponding behavioral measurement index system.Then,using K-means clustering algorithm to subdivide audiences under different user behaviors and identify their types.The results are as follows: In the audience decision-making behavior model,the audiences are subdivided into active audience,rating audience,price-based audience and negative audience;in the audience purchase behavior model,the audiences are subdivide into important value audience,important to keep audience and general development audience;in the audience post-purchase behavior model,the audiences are subdivide into high-satisfied audience,medium-satisfied audience and low-satisfied audience.Finally,targeted andpersonalized marketing strategies are proposed by analyzing different type of audience’s behavior characteristics.2.Factors related to movie marketing are selected from the entire process from movie production to release,mainly including genres,running time,ticket price,release time,score,marketing results,box office,etc.Using ridge regression,the Lasso algorithm and the Adaptive Lasso algorithm respectively establish regression models to analyze the influence of different factors on the box office.Comparing the Lasso algorithm model and the Adaptive Lasso algorithm model,the latter has better imitative effects and rationality.After combining the results,it is found that the factors to influence degree of the box office press from high to the low preface is score,amount of publicity,ticket price,genres,running time,release time.In marketing results,the greater impact on the box office in turn is amount of publicity,We Chat index and total comments.Finally,exploring the reasons why these factors have different degrees of influence on the box office and developing film marketing strategies with different focuses.3.Comprehensively analyzing the marketing data analysis results of audiences and films,and combining the development status of the film industry to summarize the problems and corresponding countermeasures of Chinese film marketing in the age of big data.At present,the marketing effect of the Chinese film industry is lagging;marketing data credibility cannot be guaranteed;the development of big data technology in film marketing needs to be considered;inadequate professionalism is still existing during the research.Therefore,in order to maximize the benefits of movie investment,it is necessary to define the film market positioning,grasp the trends of audiences and promote diversification of marketing models.Although film marketing has developed significantly,data mining cannot replace film art and artistic creation.Only maintaining a balance between these two areas can promote the vigorous development of the film industry.This research provides the necessary quantitative theoretical support in the film industry applications for the data mining technology.It provides decision-making basis for the film marketing strategy options,and enhances the profit and efficiency ofmarketing strategy.Therefore,it has a great significance in theoretical research and the practical application.