RBF神经网络非对称损失改进及应用

Improving RBF Neural Network under Asymmetric Loss and Its Application

针对应用RBF(Radial Basis Function)神经网络信用评分中存在的第Ⅰ类错误率高的问题,提出了基于Linex损失下RBF神经网络分类方法,并给出了UCI(University of California Irvine)中德国信用评分数据集上的测试结果。实验结果表明,该方法能有效解决传统RBF神经网络信用评分中存在的问题。

Aiming at the problem in the application of the RBF （ Radial Basis Function） neural network to the credit score,we propose a RBF neural network classification under the Linex loss function,and produce the test results on the Germany credit score data sets among the UCI （ University of California Irvine） . The results illuminate that the method proposed in this paper may effectively solve the problem in the application of the RBF neural network to the credit score.