基于灰色关联度的粗集连续属性离散化算法

连续属性的离散化是粗糙集理论亟待解决的关键问题之一。基于灰色系统和粗糙集的有关理论，提出了 一种新的基于属性重要性的离散化算法。该算法以条件属性对决策属性的灰色关联度来度量条件属性的重要性， 在保证决策表原始分类能力不变的前提下，按照属性重要性由小到大的顺序对每个条件属性的侯选断点进行考 察!将冗余的断点去掉，从而将条件属性离散化。同时给出了该算法的时间复杂度分析，并通过实例分析验证了算 法的有效性和实用性。

Discretization algorithm of continuous attributes in rough sets based on grey correlation degree

Discretization of continuous attributes is always one of the key problems that need urgent solutions in rough sets theory. Based on theory of grey system and rough sets, a new discretization algorithm of continuous attributes in decision table is offered. In this algorithm, the grey correlation degree of condition attributes for decision attribute is used to measure the importance of condition attributes ; on the premise of keeping the stability of original decision table, all break points of every condition attributes are examined and the redundant ones are eliminated according to which the condition attributes are sorted in a descending order, and then the decision table is discretized. The time complexity of the algorithm is proposed and an example is investigated to verify its validity and practicability.