基于动态聚类-因子分析的社会发展评价指标构建模型

发布时间：2018-04-06 03:13

本文选题：社会发展　切入点：评价指标　出处：《数学的实践与认识》2017年16期

【摘要】：运用动态聚类与因子分析相结合的双重指标筛选法对社会发展评价指标进行定量筛选,最终建立了包含社会公平,社会公共利益保障两个准则层在内的共计9个指标的社会发展评价指标体系.主要特色一是利用动态聚类的方法分别对社会公平,社会公共利益保障两个准则层内的社会发展评价指标进行分类,使不同的聚类结果代表社会发展的不同层面,避免了各个社会发展评价指标之间反映的社会发展信息重叠;二是通过因子分析遴选出每一个社会发展评价指标类中信息贡献程度最大的社会发展评价指标,同时删除其他信息贡献率小的社会发展评价指标,保证了筛选出的社会发展评价指标能最大程度的反映整个海选社会发展评价指标的信息量;三是主成分-熵的信息贡献测算模型表明:最终构建的社会发展评价指标体系用29%的社会发展评价指标覆盖了社会发展海选指标体系中89%的原始信息量.
[Abstract]:The double index screening method combined with dynamic clustering and factor analysis was used to screen the social development evaluation index quantitatively, and the social equity was finally established.Social development evaluation index system of 9 indexes, including two levels of social public interest protection.The main feature of the paper is to classify the social development evaluation indexes in the two levels of social equity and social public interest protection by using the dynamic clustering method, so that different clustering results can represent different levels of social development.Avoiding the overlap of social development information reflected among the social development evaluation indicators; second, selecting the social development evaluation indicators with the greatest contribution of information in each social development evaluation index category through factor analysis.At the same time, the social development evaluation index with small contribution rate of other information is deleted to ensure that the selected social development evaluation index can reflect the whole society development evaluation index to the greatest extent.Third, the information contribution model of principal component entropy shows that the final social development evaluation index system covers 89% of the original information in the social development selection index system with 29% of the social development evaluation index.
【作者单位】：内蒙古农业大学经济管理学院;
【基金】：中国博士后科学基金项目(2015M582754XB) 内蒙古自治区自然科学基金(2016MS0714) 内蒙古社会科学研究课题(16B12)
【分类号】：C912.4

【相似文献】