摘要
个人信用评估对于商业银行规避消费信贷风险具有重要意义。为了构建更优的个人信用评估模型,提出了利用遗传算法(GA)优化神经网络的信用评估方法,并通过GA适应度函数的设置控制信用评估中给商业银行造成损失较大的第二类误判的发生。模型的应用结果与BP神经网络进行对比表明,GA神经网络能够有效地控制第二类误判的发生,模型的稳健性高,具有更好的适用性。
Personal credit assessment is of great significance for commercial banks to avoid credit risk.For a better personal credit assessment model,this article proposes to use genetic algorithm(GA) to optimize the neural network credit evaluation and to stay off,by setting adaptive function with GA,the second category of misjudgment in credit evaluation that often causes severe loss to commercial banks.The application of the model shows that,compared with BP neural network,GA neural network is more effective in co...
出处
《武汉科技大学学报（社会科学版）》
2007年第4期368-372,共5页
Journal of Wuhan University of Science and Technology:Social Science Edition
基金
国家哲学社会科学创新基地资助项目(编号:NCTPM200408)
