论文部分内容阅读
准确的煤炭资源资产分类是有效进行煤炭资源资产管理的前提和基础.针对现有分类方法存在的不足,提出了基于人工神经网络(ANN)与粗集理论(RS)的煤炭资源资产分类方法.首先,由资产分类的样本数据形成决策表,使用专家离散法对数据进行离散处理;然后,采用遗传算法(GA)对决策表进行属性约简;最后根据约简后的属性集构建起煤炭资源资产分类的神经网络模型.实例运行表明,所提出的模型方法比单纯的ANN方法在学习效率和分类准确率方面均有所提高.
Accurate classification of coal resources assets is the prerequisite and basis for the effective management of coal resources assets.According to the existing deficiencies of the existing classification methods, a coal resource asset classification method based on artificial neural network (ANN) and rough set theory (RS) is proposed. Firstly, a decision table is formed from the sample data of the asset classification, and the data is discretized using the expert discretization method. Then, the genetic algorithm (GA) is used to reduce the attribute of the decision table. Finally, coal resources are constructed based on the reduced attribute set The neural network model of asset classification shows that the proposed model method has higher learning efficiency and classification accuracy than the simple ANN method.