Abstract: This paper studies the rough set models of combined real-valued information systems based on random set. In this paper, the knowledge structure implied in the real-valued information systems understood by expertise are represented by random sets mapping. Two kinds of combined information systems are defined formally, and their rough set models based on random set and their properties are researched in details, and the relationships between their upper and lower approximation operators are also discussed. By using the results, the upper and lower approximation operators of the combined information system can be computed from that of its original real-valued information systems. An application example is given to show the efficiency.

Key words: random sets, combined information system, rough sets, consistency, real-valued information system

摘要： 研究基于随机集的实值信息系统及其合成系统的粗糙集模型。用随机集值映射表示专家对实值信息系统知识结构的认识，给出 2类合成信息系统的形式化定义，并研究对象合成信息系统和属性合成信息系统的基于随机集值映射的粗糙集模型及其性质，讨论其中上、下近似算子之间的关系。实例证明，该模型能利用此关系由原信息系统模型估算或计算出其合成信息系统模型的上、下近似算子。

关键词: 随机集, 合成信息系统, 粗糙集, 相容性, 实值信息系统

CLC Number: 

• TP391.1