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
-
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。Abstract: 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.
下载: