This paper presents a new method to evaluate the performance of the E-government website with Neural Network technology. The evaluation model, based on BP neural network and Principal Component Analysis, raises the performance of the network. And it guarantees the objectivity and scientific of the evaluation result. This method analyzed a lot of data with Principal Component Analysis Method and the main evaluation factors were picked up. Dimensions were declined effectively on condition that evaluation information was reserved. BP neural Network was used to fix on the weights of indices. It weakens the random and appraiser?s subjectivity when fixing the index weight in evaluation. With the application and verification in E-government sites of Ningbo 2006, the model based on BP Neural Network has an extensive foreground and application value.