基于随机森林插补的缺失数据的印度肝脏病人数据支持向量机分类分析

Support vector machine classification analysis of Indian liver patient data based on missing data from random forest interpolation

通过构造不同缺失水平条件下的印度肝脏病人多重缺失数据,运用多重插补和随机森林两个插补方法补充完整数据集,并将数据集应用到支持向量机分类器当中,证明了同等条件下随机森林方法插补下的支持向量机分类器分类效果相对总优于多重插补方法下的分类效果.应用插补方法灵活地将缺失数据应用在机器学习方法当中具有很好的前景.

Multiple deletions data were constructed for Indian liver patients at different levels of deletion.Two interpolation methods,multiple interpolation and random forest were used to supplement the complete data set.The data set is applied to support vector machine(SVM)classifiers,and it is proved that the interpolation effect of the random forest method under the same conditions is better than the multiple interpolation results of SVM classification.The application of interpolation to flexibly apply missing data in machine learning has a good prospect.