基于信息论的熵值法的算法改进——以陕西省环境规制强度评价为例

Algorithm Improvement of Entropy Method Based on Information Theory——with the Assessment of Environmental Regulation Intensity in Shannxi Province as an Example

基于信息论的视角对熵值法的原理——赋权依据进行了论证,证明了指标的权重实际上是指标所传递的信息量占所有指标所传递的总信息量的比例,并基于此对传统算法进行了改进,由此选取了5个指标以改进的熵值法对陕西省2005—2013年环境规制强度进行了评价,最后说明了熵值法的局限性与权重的大小有关,并非与其自身的重要性直接相关。

From the perspective of information theory,the principle of entropy method,namely,the basis ofempowerment,isexpounded and verified. It is proved that the index weight is actually the proportion of the information conveyed by the index in all indices,based on which,the traditional algorithm is improved. Then five indices are selected to assess the environmental regulation intensity of Shannxi province from 2005 to 2013 by the improved entropy method. Finally it is explained that the limitation of entropy method is related to the size of the weight,but is not directly linked to its own importance.