摘要
针对现有指标筛选方法不能将指标客观数据和指标实际含义同时处理的弊端,研究提出了一种新型粗糙集指标筛选方法,并对绿色经济下的社会可持续发展评价指标体系构建进行了实证分析.方法同时将粗糙集决策表的相对约简理论与信息表的属性约简理论相结合,既保证了筛选的客观性又兼顾了指标的实际含义,其主要做法一是通过布尔推理算法,保证了连续型指标离散化过程中候选断点的最优组合,为粗糙集约简提供了高准确率的信息表数据;二是考虑了指标的实际含义,通过对有因果关系的指标构成的决策表进行相对约简,删除了指标信息间的冗余指标;三是通过对无实际联系的指标构成的信息表进行属性约简,删除了研究指标客观数据中的冗余指标.
In this paper, a new rough set index screening method is proposed, and the social development evaluation index system construction is analyzed empirically. In this paper, the method can be used to solve the problem of the objective index and the actual meaning of the index. This method combines the relative reduction theory of the rough set decision table with the attribute reduction theory of the information table, which not only ensures the objectivity of the screening and the actual meaning of the index. The main reason is that through Boolean algorithm, the optimal combination of candidate breakpoints in the discretization of continuous index is ensured, which provides high accuracy information table data for rough set reduction. Second, it takes into account the actual meaning of the index, the relative reduction of the decision table composed of the causal relationship index is deleted, and the redundancy index between the index information is deleted. Thirdly, the attribute data of the information table composed of the indicators without the actual connections are reduced and the objective data in the redundant indicators.
作者
刘丹
李战江
郑喜喜
LIU Dan;LI Zhan-jiang;ZHENG Xi-xi(College of Economics and Management, Inner Mongolia Agricultural University, Hohhot 010010, China)
出处
《数学的实践与认识》
北大核心
2017年第21期134-144,共11页
Mathematics in Practice and Theory
基金
中国博士后科学基金项目(2015M582746XB)
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)
关键词
指标筛选
评价指标
粗糙集
index screening
evaluation index
rough set
