粗糙认知图RCM模型研究

Research on Rough Cognitive Map Model

在继承模糊认知图模型优点的前提下,考虑到认知图和模糊认知图不能表示概念间关系的多样性及关系测度对概念状态值的动态依赖关系,也不能表示概念间因果关系测度的不确定性,将粗糙集的思想融入到认知图中,提出粗糙认知图模型,从而将概念间关系的多样性、时空特性有效的融入认知图中,并抽象出两概念间的一个综合关系,赋予权值,以完成因果关系推理,扩展了FCM的应用领域及模拟概念间关系的能力。较大限度地减少了认知图对现实世界模拟的失真。

Under the circumstance of inheriting the advantages of fuzzy cognitive map, considering that cognitive map （CM） and fuzzy cognitive map （FCM） fails to represent the diversity of the relations of concepts, the dynamic dependency between the relation measures and the state value of concepts, and the measures of uncertain causal rela-tionships ,the paper introduced the rough set to solve those problems, and rough cognitive map （RCM） is presented. RCM abstracts a comprehensive relationship between two concepts and gives a comprehensive weight to complete causal reasoning. So the simulating capability of FCM is extended. Finally, RCM is more flexible and robust than FCM, it can simulate the real world better.