判别分析与决策树分析在化学物质生态危害分类中的应用  

Application of Discriminant Analysis and Decision Tree Analysis to the Classification of Ecological Hazards of Chemicals

作　　者：周林军[1,2] 刘济宁[2] 石利利[2] 单正军[2] 陈国松[1] 蔡玉琪[2] 

机构地区：[1]南京工业大学理学院,江苏南京210009 [2]环境保护部南京环境科学研究所,江苏南京210042

出　　处：《生态与农村环境学报》2009年第1期57-61,共5页Journal of Ecology and Rural Environment

基　　金：国家重点基础研究发展计划(2002CB410800)

摘　　要：运用Fisher判别、马氏距离判别和决策树分析3种方法对61种环境优先污染物的生态危害程度进行分类,并比较了各模型的分类正确率。结果显示,决策树分析方法分类正确率最高,为92%;马氏距离判别其次,为87%;Fisher判别最低,为75%。决策树分析方法不仅减少了2项评价指标,而且对61个新数据矩阵的多次分析显示其分类能力非常稳定,正确率基本符合正态分布,且保持在92%左右,为3种方法中最优的分类方法。Decision tree analysis, Mahalanobis distance discriminant analysis and Fisher discriminant analysis were employed to classify ecological hazards of 61 chemicals. Comparison between these methods showed 92% , 87% and 75% , respectively in correctness of the classification . Decision tree was not only the highest in correctness, but also reduced 2 indexes in the evaluation. Moreover, analyses of the 61-index matrix demonstrated that the classification capacity of the decision tree method was very stable and its correctness displayed normal distribution, which remained around 92%. So it is the optimal method for the classification of ecological hazard in these methods.

关 键 词：FISHER判别 马氏距离判别 决策树 生态危害分类 

分 类 号：X826[环境科学与工程—环境工程] X11