直觉模糊概念意义下的属性约简

Attribute reduction based on intuitionistic fuzzy concepts

针对传统的概念算子的构造只考虑了对象和属性之间的一种二元关系,定义了直觉模糊信息系统.在该系统中,基于对象和属性间的直觉模糊二元正关系和二元负关系建立了关于对象集和属性集的2对算子,它们充分考虑了正、反两方面信息.通过引入精度阈值,定义了3类直觉模糊概念,并讨论了其相关性质.通过引入对象和属性间的支配关系,建立了新的概念格模型,研究了保持哈斯图结构不变的直觉模糊信息系统的属性约简.数值实例验证了该模型的有效性和实用性.

Operators which are constructed based on the lattices in traditional concept defines only one relation be- tween objects and attributes. In view of this shortcoming, an intuitionistic fuzzy information system was defined, in which two pairs of operators with regard to object set and attribute set are constructed based on a binary positive re- lation and a binary negative relation between attributes and objects, taking advantage of both the positive and nega- tive information. Three types of intuitionistic fuzzy concepts introducing the precision threshold value are defined, and their properties are discussed. A new model of conceptual lattice was derived by defining a dominance relation- ship between objects and attributes and in addition, the issue on attribute reduction of intuitionistic fuzzy informa- tion systems is investigated under the principle of keeping the Hasse graph invariant. The efficiency and practicabil- ity of this method were verified by examples of research findings.