Design and Research of Financial Risk Early Warning System Based on Cash Flow
To improve the accuracy and effectiveness of current financial risk early warning,an enterprise financial risk early warning model based on extreme learning machine(ELM)algorithm is established based on cash flow theory.The index of cash flow is screened by entropy method,and the index data are divided into test set and training set.Extreme learning machine is used for training and testing,and the adjusted model is applied to financial risk early warning.The results show that the 11 indicators extracted from the study can reflect the financial status of enterprises from many aspects.The average absolute error of the research model is the smallest,4.21％.The classification accuracy of the research model is always higher than that of the traditional algorithm model.When the number of features is 10,the classification accuracy of the research algorithm is as high as 91％.The financial early warning system designed has good performance and has certain application value.
万方数据
维普
