Research of Electronic Commercial Credit Rating Based on Neural Network with Principal Component Analysis

In order to improve the level of electronic commerce credit rating, we establish a full set of E-commerce credit rating level system, and present a method which based on the combination of a principal component analysis and BP neural networks. This method improved the traditional BP neural networks, by using a principal component analysis could eliminate the correlation between the inputs of neural network, reduce the input dimension, made the structure easy and the convergent rate of network quickly. And it can enhance the whole capability of network. The result indicates that this method can rate the E-commerce credit level quickly and effectively.