Evaluative Research into E-government Sites Based on BP Neural Network

摘要

After analyzing various evaluation index systems of E-government sites, issues were studied such as too many indices, hardly-quantified and easily- correlated indices.Principal Component Analysis Method was adopted firstly to screen out the main one from a great many evaluation factors. The dimension was declined effectively on condition that evaluation information was reserved. The E-governmental website's quantitative evaluation is gained by the self-study function of the BP neural network. With the application and verification in E-government sites of Hangzhou 2005, it has proved that the model has a good universal property, robustness and reliability. It supplies a new valid way of evaluation in E-government sites.