摘要
运用粗糙集方法和信息熵概念,按照费用特征参数相对于费用的重要度的大小,对费用特征参数集进行约简,以确定RBF神经网络输入层变量.在此基础上通过对典型样本的学习,建立装备研制费用预测模型.实例结果表明该方法具有较高的预测精度,是可行的、有效的.
In term of the important degree of the cost feature parameters to the cost, in order to set the input variables of the RBF artificial network, the Rough set approach and the conception of information entropy are employed to reduce the parameters of the cost feature parameter set. The equipment research cost estimation model is constructed by learning from the typical samples. The results show that the approach can enhance the predication accuracy, and be feasible and effective.
出处
《装备指挥技术学院学报》
2006年第5期14-17,共4页
Journal of the Academy of Equipment Command & Technology
基金
空军工程大学博士论文创新基金(YF20040032)
关键词
粗糙集
神经网络
信息熵
费用预测
Rough set
artificial network
information entropy
cost estimation
