相似粗糙集理论在CBR系统中的应用研究

The Research on the Application of Similarity Rough Set in CBR System

针对连续型属性离散化对相似度计算造成的影响，给出了一种基于相似粗糙集的属性权值计算方法，并提出基于相似粗糙集的CBR系统案例推理模型。经实例验证，该方法在先验知识不足且连续型属性众多的案例推理应用中具有很好的适用性。

Focus on the impact of continuous attributes discretization in similarity calculation,based on SRS,this paper gives out an attribute weight calculation method,puts forward a case reasoning model in CBR.This method is useful in the case with less prior knowledge and many continuous attributes.