基于二进制的知识表求核算法  

Core algorithm of knowledge table based on binary

作　　者：李天志[1] 梁家荣[2] 梁美莲[3] 范平[2] 

机构地区：[1]德州学院计算机系,山东德州253023 [2]广西大学计算机与电子信息学院,广西南宁530004 [3]广西大学数学与信息科学学院,广西南宁530004

出　　处：《计算机工程与设计》2007年第2期433-435,共3页Computer Engineering and Design

基　　金：国家自然科学基金项目(60564001)

摘　　要：求核运算是属性约简的基础,直接影响着知识约简的效率。提出了基于二进制的新的知识表求核算法,将求核运算转化为数值的比较运算,并且利用roughset理论的相关知识得出了一个重要定理,利用此定理,在求必要属性时,只需选取/IND(R)每个分类中的一个元素进行比较,大大减少了数据处理量。与传统算法相比,其速度更快,效率更高,易于计算机实现,为扩展粗糙集的应用提供了理论基础。The core obtaining plays an essential role in the attribute reduction, which affects the efficiency of attribute reduction. A new core algorithm of knowledge table based on binary is proposed. The way to get the core is transformed into the comparison of numbers. In addition, a theorem is presented based on the knowledge of rough set theory. With this theorem, only one element for each U/IND （R） is used during the process of computing the necessary attribute, So the computing data is decreased greatly. The algorithm is more efficient and effective than the traditional algorithms. And it can be realized easily with computer. It provides theoretic foundation for expanding application of rough set.

关 键 词：粗糙集 知识表 属性约简 核 二进制 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]