多尺度决策系统中基于模糊相似关系的决策粗糙集最优尺度选择与约简

Optimal scale selection and reduction of decision-theoretic rough set based on fuzzy similarity relation in multi-scale decision systems

基于模糊相似关系的决策粗糙集是经典决策粗糙集的延伸与拓展,为当今的研究热点之一。在实际处理数据时,同一对象的同一属性可能具有不同层次,并且在不同层次上取得不同值。为此,该文针对多尺度决策系统,提出多尺度决策系统中基于模糊相似关系的决策粗糙集模型,给出相应的最优尺度选择及约简方法,并讨论了获得一个最优尺度约简的简便算法。利用5组UCI标准数据集对该文所提出的模型与算法进行实例验证,证明其有效性。该文提出的模型进一步拓展了基于模糊相似关系的决策粗糙集在多尺度决策系统下的应用。

Decision-theoretic rough set based on fuzzy similarity relation is an extension of the classical decision-theoretic rough set and is one of the current hot research topics.In actual data processing,the same attribute of the same object may have different levels and get different values at different levels.A decision-theoretic rough set model based on fuzzy similarity relation in multi-scale decision systems is proposed.An optimal scale selection and reduction method is given,and a simple algorithm to obtain an optimal scale reduction is discussed.Five UCI standard data sets are used to prove the model and algorithm proposed in this paper.The model proposed in this paper further expands the application of decision-theoretic rough sets based on fuzzy similarity relation in multi-scale decision systems.