一种改进的基于差别矩阵的求核属性算法

An Algorithm for Computing Core Attribute Based on Improved Discernibility Matrix

本文研究了粗糙集中决策表的求核属性,分析相容和不相容决策表的特点。结合现有的求核属性方法,提出了适合相容和不相容决策表的基于"D集"的改进的差别矩阵求核方法。该方法的基本思想是核属性应该考虑决策属性的取值,并由取值的区域划分来决定。本文主要是对张振林等提出的算法进行改进,通过改进基于粗糙集理论的求属性核的算法,能更快,更有效地找到属性核,并通过实验验证了算法的可行性和有效性。

The core attributes in decision table of the rough set is studied and the features of compatible and incompatible decision table are analyzed. Simultaneously, combining with the existing methods for core attributes, based on the ＂D-set＂, an improved dis- cernibility matrix method that suffices the compatible and incompatible decision table is proposed. The basic idea of the method is that the core attribute should take decision attribute values into account and is determined by the values of regional division. This paper is mainly to improve the algorithm of Zhang Zhenlin. Based on the rough set theory, the algorithm is improved to get attribute core which make the process more effective and fast. At last, the feasibility and effectiveness of the algorithm is verified through the experiment.