基于非对称选择相似关系的扩充粗糙集模型

Extended rough set model based on non-symmetric alternative similarity relation

针对带"*"值的不完备决策系统,提出一种基于非对称选择相似关系的扩充粗糙集模型,通过定义"选择相似"的概念来合理地控制未知值和已知值的相似程度,克服了"*"值可与任意值相似的不足。理论分析表明,该模型不但符合人的"选择性"要求,而且具有更加合理的分类效果。实例分析进一步验证了非对称选择相似关系的优越性。

In this paper, aiming at the incomplete decision system with＂＊＂, a new non-symmetric alternative similarity relational model is proposed, and by defining the concept of alternative similarity to reasonably control the similar degree of the unknown and known values, thus having overcome the similar shortage of the＂＊＂value and the random value. Theoretical analysis shows that, the model not only meets the people’s＂selective＂requirement, but also has a more reasonable classification. The case analysis further validates the advantages of non-symmetric alternative similarity relation.