信息表属性约简新方法

New Methods for Attribute Reduction of Information Table

在研究区分能力大小的基础上建立一个用于指导信息表的绝对属性约简的粗糙集模型,研究区分能力和分类能力之间的关系,提出决策依赖区分精度概念,为指导决策表的相对属性约简提供了一个新的判据。给出区分精度、近似精度和决策依赖区分精度在属性约简过程中相互关系的研究结论,通过一组对比实验说明决策依赖区分精度比近似精度对分类能力的描述更细致客观。

A rough set model is established to supervise the absolute attribute reduction for information table on the basis of studying the separating capacity. And a novel conception is proposed, which is called discernibility quality based on decision, on the basis of exploring the relations between the ability of discernibility and classifying, and it is an important criterion to supervise the relative attribute reduction for decision table. Several related conclusions are drawn by theoretical analyses in studying discernibility quality, approximate quality and discernibility quality based on decision. Comparison experiment shows that discernibility quality based on decision is finer than approximate quality for describing the ability of classifying.