不完备目标信息系统中的可变精度粗糙集模型

文中以同时具有缺席型和遗漏型未知属性值的不完备目标信息系统为研究对象,根据特征关系,研究可变精度粗糙集的模型及其性质．可变精度粗糙集模型与原始的粗糙集模型不同,它是建立在集合多数包含的基础上的,因而该模型是基于特征关系的经典粗糙集模型的推广形式,而基于特征关系的经典粗糙集模型则是可变精度粗糙集模型的一种特殊表现形式．文中对新模型的主要性质作了阐述和证明,结果表明：在不完备目标信息系统中,新模型与原始的粗糙集模型相比具有更高的近似精度,可进行更为精确的度量．

Variable precision rough set model in incomplete objective information system

In this paper, the incomplete objective information system in which unknown attributes values were considered as not only lost but also careless conditions was deeply investigated. By using the concept of charac- teristic relation, a variable precision rough set model and its properties were studied. The proposed model was different from the traditional rough set model on the basis of characteristic relation because it was constructed by the idea of inclusion on the sets. The model was a generalized form of the traditional characteristic relation based rough set model that was a special form of the proposed rough set model. The main properties of the new model were described and proved. The results show that the new model has more approximate accuracy than the classic model in the incomplete objective information system. Moreover, it can achieve more accurate measure.