论文部分内容阅读
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.
We use the dual index screening method of dynamic clustering and factor analysis to quantitatively evaluate the social development evaluation index, and ultimately establish a total of nine indicators, including social justice, social and public interests protection criteria of social development evaluation index System.The main characteristic is to classify the social development evaluation index in the two criteria of social fairness and social public interest protection respectively by using dynamic clustering method so that different clustering results represent different levels of social development and avoid all Social development evaluation indicators reflect the overlap between social development information; the second is through factor analysis to select each of the social development evaluation indicators in the category of social contribution to the greatest degree of social development evaluation indicators, and delete other information contribution rate of small social development evaluation indicators , To ensure that the selected social development evaluation index can reflect the information of the whole evaluation index of the social development of maritime election to the largest extent. Third, the information contribution estimation model of principal component-entropy shows that the final evaluation index system of social development is constructed with 29% The social development evaluation index covers the index of marine development for social development 89% of the original amount of information in the system.