基于误差度的Rough集近似质量的分析

Analysis of Approximation Quality Based on the Degree of Errors in Rough Set

在分析Pawlak近似空间Rough集的近似质量时,常采用准确性因子α和精确性因子β;在比较两个划分的好坏时采用属性的依赖度来描述。文章提出了误差度概念,利用误差度能更好地分析Rough集的近似质量,比较分划的优劣;同时也给出了属性依赖度新的表达式。

The degree of accuracy α and the degree of precision β are usually accepted when it analyzes approximation quality in rough set,and the degree of dependency γ of the attributes is often taken to describe two distributions.In this paper,the concept of the degree of error η is introduced to analyze the approximation quality more effectively and to compare the two distributions.Meanwhile,a new representation of dependency of attributions is given.