基于数据驱动与信息增益率多指标决策模型的生活饮用水水质评估  

Multiple criteria decision-making model based on data driven and information gain ratio for drinking water quality evaluation

作　　者：李佳龙[1] 阎威武[1] 白晓慧[2] 邵惠鹤[1] 

机构地区：[1]上海交通大学电子信息与电气工程学院,上海200240 [2]上海交通大学生命科学技术学院,上海200240

出　　处：《大连理工大学学报》2016年第1期92-97,共6页Journal of Dalian University of Technology

基　　金：国家自然科学基金资助项目(60974119)

摘　　要：生活饮用水的水质情况对于居民的日常生活非常重要.建立一种基于数据驱动与信息增益率的多指标决策模型来评估生活饮用水的水质.该决策模型首先通过变异系数法对多指标系统的每个指标赋予初始权重,然后通过综合加权指数评价方法得到初始的评价结果,最后应用信息增益法根据各指标贡献度调整权重后再次评估系统.该模型不仅可以综合评估生活饮用水的水质,也可以找到对评价结果起关键作用的重要指标.通过GIS直观展示了水质分布情况,说明该决策模型的有效性.The drinking water quality is important for residents′daily life,so a multiple criteria decision-making model based on data driven and information gain ratio is set up to evaluate drinking water quality.Firstly,the initial weights of every index in multiple indices system are determined by the variation coefficient method.Secondly,the initial evaluation results are obtained by the comprehensive weighted indices evaluation method.Finally,information gain method is applied to adjust the weights of indices according to the contribution degree and evaluate the system again.The model can not only comprehensively evaluate the drinking water quality,but also find the important indices playing a key role.Water quality distribution is shown intuitively by GIS to illustrate the validity of the decision-making model.

关 键 词：生活饮用水 数据驱动 变异系数 信息增益率 GIS 

分 类 号：N94[自然科学总论—系统科学]