摘要
在灰色聚类评估中,按照最大值准则我们认定聚类对象分量最高的为所属类别.当聚类系数各向量均衡取值难以分辨或最大分量的值与其它分量的值区分度很低,根据聚类系数向量最大分量判定决策对象所属类别与对聚类系数向量整体评估所得结果可能会产生矛盾.因此,以区间重合度构建了基于白化权函数的中心点和转折点的聚核权向量组和聚核加权决策系数向量用于求解灰色两阶段模型.最后,以学科评估为例说明两阶段模型中如何构建聚核权向量组和聚核加权决策系数向量,并以此证明有效性和实用性.
Grey clustering evaluation is mainly used to evaluate which category the observed object belongs to.According to the rule of maximum value,in the grey clustering evaluation,it is obvious that the largest component of the clustering object belongs to this category.However,it is difficult to determine the ascription of decision object at the case of some main component with close value of the grey clustering.Thus,a novel two-stage model based on center points,turning points and the whitenization function is proposed by degree of coincedence,including new weight vector group of kernel clustering and weighted coefficient vector of kernel clustering for decision-making.Finally,the practical application of the new weight vector group of kernel clustering in the two-stage model is illustrated by taking the subject selection as examples.
作者
谢旭东
胡明礼
XIE Xu-dong;HU Ming-li(College of Economics and Management,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)
出处
《数学的实践与认识》
北大核心
2020年第24期55-62,共8页
Mathematics in Practice and Theory
基金
中央高校基本科研业务费青年科技创新基金((理工类),NS2017059)
国家自然科学基金(71871117)
教育部人文社科基金(18YJA630066)
中国航空基金(2017ZG52080)。
关键词
灰色聚类评估
聚核权向量组
区间重合度
两阶段模型
白化权函数
grey clustering evaluation
weight vector group of kernel clustering
degree of coincedence
two-stage model
whitenization function
