作　　者： ; ; ; ;

机构地区： 深圳大学土木工程学院

出　　处： 《大连理工大学学报》 2007年第5期729-734,共6页

摘　　要： 根据模糊聚类和粗糙集理论的基本原理,利用模糊聚类的相似关系与粗糙集可辨识矩阵的特性,结合这两种方法的互补优势,研究了将之应用于建筑业综合评价的方法和步骤.通过模糊聚类法形成粗糙集下的决策信息系统,采用基于可辨识矩阵的启发式约简法对属性约简,计算出约简后各属性的权重.利用各属性的权重分别对各省市(或地区)的建筑业指标评价值加权求和从而得出各自的建筑业综合评价值,对其结果的客观性和实用性所进行的分析评价证明该方法有效. According to the basic principles of fuzzy aggregation and rough set theory, the similar relationship of fuzzy aggregation was combined with the characteristics of discernibility matrix of rough set and the application of fuzzy aggregation and rough set theory was studied in comprehensive evaluation of building industry based on their complementary advantages. The decision information system was formed through fuzzy aggregation,then the heuristic algorithm of attribute reduction based on discernibility matrix was used to decide the sequence of attribute reduction and ascertain attribute weight. According to the weight of each attribute, the attribute values of building industry in each province or region were summed to get an integrated value. Through analyzing and evaluating the objectivity and practicality of the results, the validation of the evaluating method used is demonstrated.

关 键 词： 模糊聚类 模糊相似关系 粗糙集 可辨识矩阵 建筑业评价

领　　域： [经济管理]