一种改进的启发式离散化算法及应用  

An Improved Heuristic Algorithm for Discretization and Its Application

作　　者：彭佳文[1] 

机构地区：[1]广西工学院信息网络中心,广西柳州545006

出　　处：《计算机与现代化》2008年第9期51-53,57,共4页Computer and Modernization

基　　金：湖南省科技厅科技计划项目(05JT1013)

摘　　要：Nguyen S.H提出的布尔逻辑和粗糙集理论相结合的离散化算法是粗糙集理论中的离散化算法在思想上的重大突破。通过定义分界点来区分Nguyen S.H离散化算法中定义的断点对决策系统的分辨关系是否有贡献,并仅取分界点集作为初始断点集,使得初始断点数目较大幅度地降低,提出了一种改进的启发式离散化算法并应用于一个实际的决策系统的连续属性离散化。应用实例表明改进算法较大程度地减小了算法空间复杂性和时间复杂性,具有正确性和实用性。The discretization algorithm of rough set and boolean reasoning approach presented by Nguyen S. H is a momentous breakthrough of thinking in the diseretization algorithm of rough set theory. By defining dividing point, we can differentiate each of the breaking points defined by Nguyen S. H discretization algorithm whether has contribution to the differentiation relation of the decision system. And by choosing the dividing point set as the original breaking point set only, we can decrease the number of original breaking point to a biggish degree. This paper introduces an improved heuristic algorithm for discretization and applies it to discretization of continuous attributes of an actual decision system. The application case indicates that the improved algorithm can reduce preferably the space complexity and time complexity of the discretization and has the correctness and practicability.

关 键 词：粗糙集 离散化 决策系统 分界点 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]