Research On Influencing Factors And Prediction Of Crowd-funding Performance In Cultural And Creative Field

Culture creative field with high knowledge,high value-added and high integration characteristics of the micro cultural creativity,small and medium enterprises and individual as the principal part of its industry is faced with traditional channels of financing difficulties,financing your problems,and the raise of the subject provides a new mode of financing,help to efficient allocation of market resources.Based on the data of Modian.com,a crowd-funding platform in the cultural and creative field,this paper takes crowd-funding performance as the main research objective and systematically analyzes the influencing factors and forecasting methods of crowd-funding performance.This paper first constructs a multiple regression model to explore the influencing factors of crowdfunding performance.Different from historical literature,this paper adds variables such as the credit characteristics of sponsors and fincing term,so as to improve the evaluation of project quality and risk.The results show that the number of comments and supporters of investor behavior,the number of project updates,the number of return grades,the number of sponsors’ fans,and the minimum amount of support in project risk have positive effects on financing performance,while the financing term and target financing amount in risk have opposite effects.In addition,this paper divides several different project types(such as design,tide play model,etc.),deeply explores the difference of influencing factors of different types of crowdfunding performance,and discusses the moderating effect of project types on regression results.Secondly,BP neural network prediction model of crowd-funding project financing performance is constructed.Because the traditional model has the defect of falling into local optimum easily,this paper introduces the genetic algorithm to optimize it innovatively,and analyzes the establishment steps of the optimized GA-BP model and its global optimization method in detail.The empirical results show that the optimization model can effectively improve the prediction accuracy.