多类类别不平衡学习算法:EasyEnsemble.M  

EasyEnsemble. M for Multiclass Imbalance Problem

作　　者：李倩倩[1] 刘胥影[1,2] 

机构地区：[1]东南大学计算机科学与工程学院计算机网络和信息集成教育部重点实验室,南京211189 [2]南京大学计算机软件新技术国家重点实验室,南京211189

出　　处：《模式识别与人工智能》2014年第2期187-192,共6页Pattern Recognition and Artificial Intelligence

基　　金：国家自然科学基金青年基金项目(No.61105046);教育部高等学校博士学科点专项科研基金项目(No.20110092120029);南京大学软件新技术国家重点实验室开放课题项目(No.KFKT2011B01)资助

摘　　要：随机欠采样方法忽略潜在有用的大类样本信息,在面对多类分类问题时更为突出.文中提出多类类别不平衡学习算法:EasyEnsemble.M.该算法通过多次针对大类样本随机采样,充分利用被随机欠采样方法忽略的潜在有用的大类样本,学习多个子分类器,利用混合的集成技术最终得到性能较优的强分类器.实验结果表明,与常用的多类类别不平衡学习算法相比,EasyEnsemble.M可有效提高分类器的G-mean值.The potential useful information in the majority class is ignored by stochastic under-sampling. When under-sampling is applied to multi-class imbalance problem, this situation becomes even worse. In this paper, EasyEnsemble. M for multi-class imbalance problem is proposed. The potential useful information contained in the majority classes which is ignored is explored by stochastic sampling the majority classes for multiple times. Then, sub-classifiers are learned and a strong classifier is obtained by using hybrid ensemble techniques. Experimental results show that EasyEnsemble. M is superior to other frequently used multi-class imbalance learning methods when G-mean is used as performance measure.

关 键 词：机器学习 类别不平衡学习 欠采样 集成 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]