论文部分内容阅读
根据改进的层次分析法所确定的数字资源服务绩效评价体系中各评价指标的权重,利用岭型函数建立了评价指标对评价等级的隶属度函数。从而为数字资源服务绩效的评价体系进行模糊综合评价,并求出神经网络的训练样本,最后将训练好的BP神经网络对数字资源服务绩效进行评价。评价结果表明神经网络模型评价与模糊综合模型评价的结果是一样的,这也说明文中构建的评价指标体系是合理的、有效的。
According to the weight of each evaluation index in the digital resource service performance evaluation system determined by the improved AHP, the membership function of the evaluation index to the evaluation grade is established by using the ridge function. So as to make a fuzzy comprehensive evaluation for the evaluation system of digital resource service performance and to find the training samples of neural network. Finally, the trained BP neural network is used to evaluate the performance of digital resource services. The evaluation results show that the neural network model evaluation and the fuzzy comprehensive model evaluation result are the same, which also shows that the evaluation index system constructed in this paper is reasonable and effective.