一种基于灰色绝对关联度的变精度粗糙集模型

摘要

应用经典粗糙集理论处理连续值属性决策表问题时，对连续值属性进行离散化会造成信息损失，同时经典粗糙集模型也无法处理不精确分类。针对以上问题，建立了基于灰色绝对关联度的变精度粗糙集模型，研究了模型的部分性质和定理，提出了基于重要度的约简算法对模型进行属性约简。通过实例说明了模型建立和约简的具体过程。%When classical rou tribute decision table, discretizi set theory is used to deal with the problems of continuous valued at- ng continuous valued attributes will result in loss of information. Meanwhile the classical rough set model can not deal with imprecise classification. To solve the above problems, variable precision rough set model based on the gray absolute correlation degree has been established and some properties of the model are studied ; and then a reduction algorithm based on the significant degree is designed; finally, an example is used to illustrate the process of the model foun- dation and reduction.