【摘要】 文章应用神经网络数据分析技术研究ETC系统客户流失的状况,详细研究了如何建立ETC系统客户流失基本模型。通过对客户的基本数据进行神经网络预测,可以发现描述流失客户基本特征的属性值集合以及对应的是否流失的结论。文章给出的是改进的神经网络的预测方法,可提高BP神经网络的收敛速度,增强网络的泛化能力,获得了很好的效果。更多还原

【Abstract】 The authors research churn analysis of ETC system customer based on neural network data analysis technology in this paper, and describe particularly how to establish basic churn model of ETC system customer. The authors can find attribute set of churn customer’s basis character and the conclusion that whether or not churn by implementing neural network forecast on the basis data of customer. The authors adopt a modified forecast method of neural network, which improves convergent speed, strengthens extensive ability of BP neural network and gains a good effect.更多还原