论文部分内容阅读
利用随机森林法(RF)进行分类和回归,在过去被广泛地研究,然而在基于顺序响应的情况下并没有标准的方法.在随机森林(RF)的基础上通过广泛的研究,对条件推理树进行探索,以探讨结合顺序信息是否改善预测性能或提高变量选择的效果.本文提出的两种置换变量的重要性预测方法RPS-VIM和MAE-VIM经过实验验证是优化的方法,它替代目前存在的内置变量重要性测量方法ER-VIM和MSE-VIM.基于真实数据的结果表明在某些设置中,有序回归树中使用RPS-VIM和MAE-VIM顺序响应的组合,预测的排名可以得到改善,并且预测精度优于原始的基于分类树的随机森林.“,”Random forest method (RF) is used for classification and regression,which has been widely studied in the past.However,there is no standard method in the case of sequential response.On the basis of random forest (RF),through extensive research,the conditional inference tree is explored to explore whether the combination of sequential information to produce the effect of improving the predictive performance or variable selection.The importance of RPS-VIM and MAE-VIM in the prediction of the importance of the two replacement variables is reasonable,and it replaces the current existence of the ordinal response variables of importance measurement ER-VIM and MSE-VIM.The results based on real data show that in some settings,the combination of RPS-VIM and MAE-VIM sequential response in the ordinal regression trees can be improved,and the prediction accuracy is better than that of the original random forest based on the classification tree.