Research on Credit Risk Classification Model of SMEs Based on Ensembled Feature Selection
Taking the customer default rate as the evaluation standard of SMEs'credit risk,this pa-per attempts to construct a credit risk classification model for SMEs based on integrated feature selec-tion,and analyzes the model by combining mutual information matrix,random forest based on cross validation and support vector machine.The results show that the reputation rating,output efficiency and the highest output of enterprises have significant effects on credit risk,while other factors have no significant effects on credit risk,Furthermore,the experiment shows that the support vector ma-chine based on cross validation has a reliable credit risk prediction ability,and has a strong refer-ence value for micro credit risk assessment of SMEs.
万方数据
维普
