摘要
针对灰色聚类评估过程中权重的确定问题,在聚类系数离差平方和最大化的目标下,研究了灰色聚类评估模型的优化问题.通过指标白化权函数来反映聚类指标信息的差异度,提出一种基于指标白化权函数值离差平方和最大化的聚类指标权重优化方法.结果表明,优化的灰色聚类评估模型能够更好地区分聚类信息,使得聚类评估模型的意义更加显著.最后通过实例验证了该方法的有效性和优越性.
Aiming at the determination issues of indicator weights in grey cluster,this paper the sum of squares and maximum goal,and accordingly determined the weights of classification indicators.On this basis,this paper proposed grey clusters with variable weights.The results show that this method can well integrate experience of the experts and the information of a cluster sample and can be applied in the situation that indicators for clustering have different meanings,dimensions,and sizes of observational data.Finally,an example is studied to demonstrate the validity and superiority.
出处
《内蒙古师范大学学报（自然科学汉文版）》
CAS
北大核心
2012年第5期462-466,473,共6页
Journal of Inner Mongolia Normal University(Natural Science Edition)
基金
国家社会科学基金资助项目(10BFX071
12BGL064)
关键词
灰色聚类评估
权重
离差平方和
优化
grey cluster evaluation
weight
sum of deviation squares
optimization
