基于可变精度粗糙集的决策规则格构造  

The Construction of Decision Rule Lattice Based on Variable Precision Rough Set

作　　者：谷雨[1,2] 

机构地区：[1]西安交通大学电子与信息工程学院 [2]云南民族大学教育技术学院,云南昆明650031

出　　处：《微电子学与计算机》2005年第4期174-176,180,共4页Microelectronics & Computer

摘　　要：基于概念格理论提出了决策规则格的概念,格中的每一节点均可表示为相应的决策规则。针对格节点之间存在偏序关系的特点,文中给出了三种剪枝策略。特别地,将粗糙集知识约简的思想和可变精度粗糙集的β下近似思想应用于决策规则格的构造,有效地减少了格中节点的数目,并有较强的抗噪声数据干扰能力。This paper proposed a new concept of decision rule lattice based up on the theory of concept lattice. Each node of the decision rule lattice can be represented as the corresponding decision rule. Three different pruning strategies to deal with the partial order relationship among nodes are advanced. In particular, we applied the ideas of knowledge reduction and β low approximation in variable precision rough set on the construction of decision rule lattice. Therefore, the strategies can effectively reduce the number of node in lattice and have a robust capability of anti noisy data.

关 键 词：决策规则格 可变精度粗糙集 概念格 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]