一种基于属性核的Rough集属性约简搜索算法

本文提出了一种新的可以快速寻找Rough集约简属性的算法,该算法利用计算属性核来提高计算效率.首先,从属性核出发,不断地在属性核上添加非核属性以构成新的属性集合,并判断新的属性集合是否为属性约简,然后在此基础上进行剪枝运算并在不构成约简的属性集合的基础上再生成新的属性集合并进行属性约简判断.该算法为一种搜索算法,由于采用了属性核算法和剪枝算法,提高了计算效率并能够找到给定信息系统的全部属性约简.通过测试,表明该算法是一种可行且有效的属性约简算法.

Study on Searching Algorithm for Attribute Reduction of Rough Sets Based on Core Attributes

The aim of this paper is to propose an efficient algorithm for attribute reduction of rough sets. The concept of core attributes was adopted to make this algorithm more efficient. Firstly, each at- tribute was added to core attributes one by one and new attribute sets containing core attributes and at- tributes not being core attribute was generated. Secondly, whether these new attribute sets were reduced at- tribute sets was judged. Thirdly, on the basis of judgment results, new attribute sets were generat...