基于相关分析-粗糙集理论的生态评价指标体系构建

Establishment of ecological evaluation indicators system based on correlation analysis-rough set theory

根据"坚持以人为本,树立全面、协调、可持续的发展观,促进经济社会和人的全面发展"的科学发展观的内涵,以DPSIR五因素模型为概念框架,利用相关分析与粗糙集相结合的方法定量筛选指标,构建了生态评价指标体系.主要创新与特色:一是通过采用DPSIR概念框架构建指标体系,将人的需求、经济发展和社会进步等要素纳入到生态系统评价中,克服了现有生态评价只片面关注环境与资源状况或者只关注环境保护与治理状况的不足.二是通过相关性分析删除同一准则层内相关系数大的指标,避免了指标反映信息重复.三是通过模糊粗糙集删除同一准则层内给定精度下不影响评价对象分类的指标,保证了筛选出的指标体系对评价结果有显著影响.

According to ＂adhere to people-oriented, establishing a comprehensive, coordinated and sustainable development and promoting economic, social and human all-round development＂ of the Scientific Outlook on Development, using the driving force-pressure-state-impact-response （DPSIR） model, this paper constructs an ecosystem evaluation index system through quantitative approaches of correlation and rough set. The contri- bution lies in： firstly, the paper constructs an index system based on DPSIR model, which contains human demand, economic development and social progress. The system is better than the existing ones which only considers the status of environment and resources or the status of environmental protection and pollution treat- ment. Secondly, it uses correlation analysis to delete the indicators with high correlation in the same criteria layer, which avoids the duplication of information of indicators. Thirdly, under the given precision, the paper deletes the indicators with no significant impact on evaluation results through the rough set theory in the same criteria layer, which ensures that the selected indicators have significant effect on evaluation results.