区域生态安全评价的熵组合权重属性识别模型  

Attribute Recognition Method of Regional Ecological Security Evaluation Based on Combined Weight on Principle of Relative Entropy

作　　者：吴开亚[1] 金菊良[2] 

机构地区：[1]复旦大学公共管理与公共政策创新基地,上海200433 [2]合肥工业大学土木与水利工程学院,安徽合肥230009

出　　处：《地理科学》2008年第6期754-758,共5页Scientia Geographica Sinica

基　　金：国家自然科学基金项目(50579009);安徽省自然科学基金项目(050450303)资助

摘　　要：为合理地确定区域生态安全评价的指标权重,提出把指标的权重结构分为反映评价指标不同属性对生态安全等级的影响程度的主观权重,和反映各区域评价指标样本值差异信息对生态安全等级的影响程度的客观权重,可分别采用层次分析法和熵权法确定这些权重,再用最小相对信息熵原理把它们综合为组合权重;为体现单指标评价过程中的评价作用,提出用属性识别模型进行单指标评价的新思路;对组合权重值和单指标评价值进行相乘并累加,建立了区域生态安全评价的熵组合权重属性识别模型(AR-CWE)。结果说明:用AR-CWE既可利用指标中的专家主观经验信息,又可挖掘各区域评价指标样本值的客观差异信息,权重信息利用全面,评价结果合理,方法通用,在安全系统综合评价中具有一定的应用价值。In order to determine indexes weights in regional cal security evaluation, a weight structure of evaluation indexes consisted of subjective weight and objective weight was presented. The two weights can be computed by using Analytic Hierarchy Process and entropy weight method, and combination weight can be synthesized according to minimum relative information entropy principle. In order to realize the evaluation function of single index evaluation process, a new evaluation method of single index was proposed using attribute recognition model. And then a new model for evaluating regional ecological security, named AR-CWE for short, was established. The result shows that both subjective information of expert experience and objective variation information of samples values can be fully mixed using AR-CWE, and its evaluated results are reasonable. AR-CWE is a general and can be applied to comprehensive evaluation of different security systems.

关 键 词：区域生态安全评价 指标权重 单指标评价 信息熵方法 属性识别模型 

分 类 号：F830[经济管理—金融学]