为了从有偏好信息但信息不完全的多属性决策系统中获取概率决策规则,提出一种新的不完全信息的多属性粗糙决策分析方法。首先,提出扩展优势关系下相容度的概念;其次,基于相容度给出知识的粗糙近似.并证明了粗糙近似的基本性质;再次,给出粗糙近似的分类质量与β-约简的概念,并从不完全信息的偏好决策系统中导出概率决策规则;最后,通过一个实例说明新方法的可行性和有效性。
In order to discover probabilistic decision rules in preferential multiple attribute decision system with incomplete information,an extension of the rough sets model is proposed in the paper.Firstly,the concept of consistency degree based on extended dominance relation is presented;Secondly,rough approximations of knowledge based on consistency degree are defined and basic properties of rough approximations are proved.Thirdly,the classification quality of rough approximations andβ-educe of knowledge are discussed and the probabilistic sorting decision rules are given.Finally,the feasibility and effectiveness of the method are demonstrated by a real example.