基于动态聚类-因子分析的社会发展评价指标构建模型

Construction of Social Development Evaluation Index System Model Based on Dynamic Clustering-Factor Analysis

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.

The social development evaluation index system including a total of 9 indicators and two rule layers was set up by using the method of dynamic clustering and factor analysis. The main feature one is using the method of dynamic clustering respectively to classify the evaluation indexes from the social justice and social public interests protection layers; Two is selecting the social development evaluation index which has the greatest information contribution in each social development evaluation index class by factor analysis, at the same time removing the social development evaluation index with little contribution rate of other information; Three is the main component - information entropy contribution calculation model shows that the index system constructed with 29% of social development evaluation index has covered 89% of the original information volume.