论文部分内容阅读
针对粗糙集理论中基于差别矩阵的属性约简方法存在的不足,提出一种基于幂图的属性约简算法.首先通过修改样本决策属性值将不相容决策表转化为简化的相容决策表;然后将样本对概念与幂图概念相结合,将基于修正差别矩阵的不相容决策表的属性约简转化为幂图的搜索问题;最后通过实例和实验验证了所提出算法的有效性.
Aiming at the deficiencies of attribute reduction based on discernibility matrix in rough set theory, this paper proposes an attribute reduction algorithm based on power graph.Firstly, by changing the attribute value of sample decision, the incompatible decision table is transformed into a simplified compatible decision table Then, the concept of the sample is combined with the concept of the power graph, and the attribute reduction of the incompatible decision table based on the modified difference matrix is transformed into the search problem of the power graph. Finally, the validity of the proposed algorithm is verified by examples and experiments.