基于粗集的空间对象分类学习算法  

Rough-set based algorithm on learning classification rules of spatial objects

作　　者：王儒敬[1] 葛运健[2] 滕明贵[2] 张晓明[2] 

机构地区：[1]中国科学技术大学自动化系,安徽合肥230026 [2]中国科学院合肥智能机械研究所,安徽合肥230031

出　　处：《中国科学技术大学学报》2006年第2期163-169,共7页JUSTC

基　　金：国家自然科学基金重点项目(50534050)资助

摘　　要：针对二维空间对象分类问题,提出将粗集方法应用于分类规则的学习.首先建立空间分类信息表,然后采用粗集方法约简条件属性、提取分类规则,并采用多种指标对规则进行评价.通过实验结果分析,证明粗集方法能够有效用于学习空间对象分类规则.A method that uses rough set for the learning of classification rules about 2-dimensional spatial objects was presented. First, an information table of spatial classification was created. Then, rough set was used to reduce condition attributes and extract classification rules from the table. The extracted rules were evaluated according to several criteria. Experiments demonstrate that the method of rough set is efficient for learning classification rules of spatial objects.

关 键 词：空间对象 粗集 分类规则 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]