Exploratory Undersampling for Class-Imbalance Learning
Exploratory Undersampling for Class-Imbalance Learning
复制标题
DOI：
10.1109/tsmcb.2008.2007853
复制
发表时间：
2009-04-01
影响因子：
--
通讯作者：
Zhou, Zhi-Hua
中科院分区：
文献类型：
--
作者：
Liu, Xu-Ying;Wu, Jianxin;Zhou, Zhi-Hua
欠采样是处理类不平衡问题的一种流行方法，它只使用多数类的一个子集，因此是非常有效的。主要的不足是忽略了许多大多数类的例子。为了克服这一缺陷，我们提出了两种算法。EasyEnSemble从多数类中采样几个子集，使用每个子集训练学习者，并组合这些学习者的输出。BalanceCascade按顺序训练学习者，其中在每一步中，由当前训练的学习者正确分类的大多数类别示例被从进一步的考虑中移除。实验结果表明，这两种方法的ROC曲线下面积、F-度量和G-均值都高于现有的许多类不平衡学习方法。此外，当使用相同数量的弱分类器时，它们的训练时间与欠采样方法大致相同，明显快于其他方法。
Undersampling is a popular method in dealing with class-imbalance problems, which uses only a subset of the majority class and thus is very efficient. The main deficiency is that many majority class examples are ignored. We propose two algorithms to overcome this deficiency. EasyEnsemble samples several subsets from the majority class, trains a learner using each of them, and combines the outputs of those learners. BalanceCascade trains the learners sequentially, where in each step, the majority class examples that are correctly classified by the current trained learners are removed from further consideration. Experimental results show that both methods have higher Area Under the ROC Curve, F-measure, and G-mean values than many existing class-imbalance learning methods. Moreover, they have approximately the same training time as that of undersampling when the same number of weak classifiers is used, which is significantly faster than other methods.