摘要
基于粗糙集的属性约简过程中,属性集对论域的划分存在大量繁杂且冗余的集合求交运算,提出一种将求交集的划分过程变换为对一组一维整数集合进行分类的替换算法.通过一个典型的算例说明该算法的合理性和有效性.与传统的算法比较,该算法具有简明、易实现的特点.
There exists a lot of complex and redundant set intersection operations in the process of attribute reduction based on the rough set. This paper presents an improved algorithm which uses the method of partitioning a group of one-dimension integers for domain partition instead of set intersection operations. A representative example illustrates that the algorithm is feasible. The given algorithm is featured by its simplicity and easy implementation, compared with the conventional algorithm.
出处
《空军雷达学院学报》
2004年第3期57-60,共4页
Journal of Air Force Radar Academy
关键词
粗糙集
数据开采
属性约简
rough set
data mining
attribute reduction
作者简介
徐新(1980-),男,硕士生,主要从事人工智能和数据挖掘研究.
