Abstract: For representing the uncertain knowledge with the characteristics of incomplete and multi-granularity, a framework of rough description logics named Rough-SHOIN is designed. Rough similarity relation is used as the basis of concept description for defining rough upper and lower approximation of concepts to represent the incomplete notions. Context is introduced to give an exact definition of concept-object relationship for different granularity in concept interpretation. Based on above technologies, syntax, semantics and knowledge base of Rough-SHOIN are defined, and reasoning rules are also expatiated without increasing calculating complexity. Examples are introduced to testify the validity of this framework.

Key words: knowledge representation, uncertain concepts, description logics, rough similarity relation

摘要： 针对具有不完备、多粒度特点的不确定知识表示，设计了一个粗糙描述逻辑框架——Rough-SHOIN，利用粗糙相似关系作为概念描述的基础，定义概念的粗糙上近似和下近似，实现不完备概念表达，在概念解释中引入上下文，实现在不同粒度上准确定义概念-对象间关系。在此基础上，给出Rough-SHOIN粗糙概念的语法、语义及知识库，并在不增加计算复杂度的前提下，定义Rough-SHOIN的推理规则。通过实例说明该框架的有效性。

关键词: 知识表示, 不确定概念, 描述逻辑, 粗糙相似关系