基于灰色动态聚类-粗糙集的绿色资源环境评价指标构建模型

发布时间：2018-03-10 05:06

本文选题：绿色资源环境　切入点：评价指标　出处：《统计与信息论坛》2017年08期 　论文类型：期刊论文

【摘要】：以内蒙古自治区12个盟市的绿色资源环境发展为研究对象,采用灰色动态聚类与粗糙集相结合的方法,构建包含有全年供水量等11个指标的内蒙古自治区绿色资源环境指标体系,其要点在于:一是通过灰色关联分析建立样本间的灰色关联矩阵,进而进行样本间的灰色聚类,反映样本间的信息重复性;二是采用动态聚类方法,每次去除一个指标后继续通过灰色关联分析建立的灰色关联矩阵进行灰色样本聚类,为粗糙集约简提供信息数据;三是通过粗糙集约简理论判断海选指标对聚类结果的影响是否显著,将每一次的聚类结果与原始聚类结果比较,保留两次聚类结果不同且对评价样本分类有显著影响的海选指标;四是采用非参数Kruska-Wallis检验的P值检验法证明本文构建的指标体系是合理的。通过对比分析表明,本文的灰色动态聚类-粗糙集指标筛选模型优于现有研究的聚类-灰色关联指标筛选模型。
[Abstract]:Taking the development of green resources and environment in 12 cities of Inner Mongolia Autonomous region as the research object, the method of combining grey dynamic clustering with rough set is adopted. A green resource environment index system of Inner Mongolia Autonomous region including 11 indexes such as annual water supply is constructed. The main points of this system are as follows: first, the grey correlation matrix among samples is established through grey correlation analysis, and then the grey clustering among samples is carried out. It reflects the information repeatability among samples. Secondly, it adopts dynamic clustering method. After removing one index, grey correlation matrix is used to cluster grey samples, which can provide information data for rough set reduction. Third, the rough set reduction theory is used to determine whether the impact of the selection index on the clustering results is significant, and the clustering results of each time are compared with the original clustering results. Keeping the two different clustering results and having a significant impact on the classification of evaluation samples. Fourthly, the P value test method of non-parametric Kruska-Wallis test is used to prove that the index system constructed in this paper is reasonable, and the comparative analysis shows that the index system constructed in this paper is reasonable. The grey dynamic clustering-rough set index screening model is superior to the existing clustering-grey relational index screening model.
【作者单位】：内蒙古农业大学理学院;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《基于小样本的内蒙古微型企业的信用评价研究》(2015M582746XB) 内蒙古自治区自然科学基金《基于违约状态判别的内蒙古微型企业的信用评价模型研究》(2016MS0714) 内蒙古社会科学研究课题《内蒙古绿色经济的评价体系研究》(16B12)
【分类号】：F062.1

【相似文献】