中文摘要：

本文在传统模糊神经网络基础上，采用灰狼优化算法计算神经网络的初始权值和阈值，提出了一种改进型模糊神经网络算法，并建立了信用卡客户违约预测模型。改进型模糊神经网络具有很好的非线性拟合能力和很好的全局搜索能力，解决了传统模糊神经网络算法收敛速度慢，容易陷入局部最优的问题。最后，通过预测信用卡客户违约问题，与支持向量机算法、传统模糊神经网络算法和卡方自动交互诊断器算法相比较，验证了改进型模糊神经网络算法的准确性、高效性和鲁棒性，平均准确率达到了94．1％。

英文摘要：

This paper was based on the traditional fuzzy neural network, which applied the gray wolf optimization algorithm to calculate initial weights and thresholds of neural network, it proposed improved fuzzy neural network algorithm and established a prediction model of default of credit card clients. Improved fuzzy neural network has good nonlinear fitting ability and global search capability and solved the problem that the traditional fuzzy neural network algorithm slowly convergence and easily falls into local solution. Finally comparing support vector machine algorithm, the traditional fuzzy neural network algorithm and Chi-square automatic interaction diagnosis algorithm with improved fuzzy neural network by prediction of default of credit card customers and verified the accuracy, efficiency and robustness of the improved fuzzy neural network algorithm, the average accuracy rate of improved fuzzy neural network was 94.1%.