基于神经网络的政府开放数据网站评价研究——以美国20个政府开放数据网站为例

Evaluation about Government Open Data Website Based on Neural Network——Taking 20-American Government Open Data Websites for Example

本文基于对政府网站评价和政府开放数据评价的研究,建立了政府开放数据网站评价指标体系。以美国20个具有代表性的政府开放数据网站为例进行指标数据获取和专家打分,在此基础上利用神经网络的BP算法确定总体的评价结果,与专家打分结果进行比较,满足误差精度的要求,从而验证了该方法应用于评价政府开放数据网站的可行性,为今后评价该类网站提供了科学、迅速、精准的评价体系。

This paper built an index system for evaluating the government open data website based on the studies abut government website evaluation and govermnent open data evaluation. By taking 20 American representative goverruneutal open data websites as a sample, this paper got the index data and score by experts and obtained the overall evaluation results through BP algorithm of neural network to compare the experts＇ scores for getting the requirement of error precision. So it could test and verify the feasibility of this way to evaluate the government open data website and could supply a scientific, rapid and accurate system to evaluate the government open data website in the future.