武森等：基于聚类欠采样的集成不均衡数据分类算法 ·1253· 参考文献 (李江，金辉，刘伟.基于分形SMOTE重采样集成算法圈定 [1]Napierala K.Stefanowski J.Types of minority class examples and 区域化探异常.计算机应用研究，2012,29(10)：3744) their influence on leaming classifiers from imbalanced data.JIn- [9]Liu X Y,Wu J X,Zhou Z H.Exploratory under-sampling for tell1 nf Syst,2016,46(3):563 class-imbalance learning.IEEE Trans Syst Man Cybernetics Part B [2]Glauner P,Boechat A,Dolberg L,et al.Large-scale detection of Cybernetics,2009,39(2):539 non-technical losses in imbalanced data sets /2016 IEEE Power [10]Mani I,Zhang I.kNN approach to unbalanced data distribu- Energy Society Innovative Smart Grid Technologies Conference tions:a case study involving information extraction /Proceed- (ISGT).Minneapolis,2016 ings of the ICML 2003 Workshop on Learning from Imbalanced [3]Haque M N,Noman N.Berretta R,et al.Heterogeneous ensem- Datasets.Washington DC,2003:42 ble combination search using genetic algorithm for class imbal- [11]Kubat M,Matwin S.Addressing the curse of imbalanced training anced data classification.Plos One,2016,11(1):e0146116 sets:one-sided selection /International Conference on Machine [4]Klein K,Hennig S,Paul S K.A bayesian modelling approach Learning.Scotland,2012:179 with balancing informative prior for analysing imbalanced data. [12]Zhu Y Q,Deng W B.A method using clustering and sampling Plas0ne,2016,11(4):c0152700 approach for imbalance data.Nanjing Unir Nat Sci,2015,51 [5]Chawla N V,Bowyer K W,Hall LO,et al.SMOTE:synthetic (2):421 minority over-sampling technique.J Artif Intell Res,2002,16: (朱亚奇，邓维斌.一种基于不平衡数据的聚类抽样方法 321 南京大学学报（自然科学版），2015,51(2)：421) [6]Zhang Y,Li Z R,Liu X D.Active learning SMOTE based imbal- [13]Dietterich T G.Machine leaming research:four current direc anced data classification.Comput Appl Software,2012,29 (3): tions.Artif Intell Mag,1997,18(4):97 91 [14]Harrington P.Translated by Qu Y D.Li R,Wang B,et al.Ma- (张永，李卓然，刘小丹.基于主动学习SMOTE的非均衡数 chine Learning in Action.Beijing:People's Posts and Telecom- 据分类.计算机应用与软件，2012,29(3)：91) munications Press,2013 [7]Wang C X,Zhang T,Ma C S.Improved SMOTE algorithm for im- (Peter Harrington.曲亚东，李锐，王斌，等译.机器学习实 balanced datasets.J Frontiers Comput Sci Technol,2014,8(6): 战.北京：人民邮电出版社，2013) 727 [15]Li Y J,Guo H X,Li Y N,et al.A boosting based ensemble (王超学，张涛，马春森.面向不平衡数据集的改进型SMOTE learning algorithm in imbalanced data classification.Syst Eng 算法.计算机科学与探索，2014,8(6)：727) Theory Practice,2016,36(1):189 [8]Li J,Jin H,Liu W.Novel improved SMOTE resampling integrat- (李诒靖，郭海湘，李亚楠，等。一种基于Boosting的集成学 ed algorithm based on fractal for geochemical anomalies evalua- 习算法在不均衡数据中的分类.系统工程理论与实践， tion.Appl Res Comput,2012,29(10):3744 2016,36(1):189)武 森等: 基于聚类欠采样的集成不均衡数据分类算法 参 考 文 献 [1] Napierala K, Stefanowski J. Types of minority class examples and their influence on learning classifiers from imbalanced data. J In鄄 tell Inf Syst, 2016, 46(3): 563 [2] Glauner P, Boechat A, Dolberg L, et al. Large鄄scale detection of non鄄technical losses in imbalanced data sets / / 2016 IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT). Minneapolis, 2016 [3] Haque M N, Noman N, Berretta R, et al. Heterogeneous ensem鄄 ble combination search using genetic algorithm for class imbal鄄 anced data classification. Plos One, 2016, 11(1): e0146116 [4] Klein K, Hennig S, Paul S K. A bayesian modelling approach with balancing informative prior for analysing imbalanced data. Plos One, 2016, 11(4): e0152700 [5] Chawla N V, Bowyer K W, Hall L O, et al. SMOTE: synthetic minority over鄄sampling technique. J Artif Intell Res, 2002, 16: 321 [6] Zhang Y, Li Z R, Liu X D. Active learning SMOTE based imbal鄄 anced data classification. Comput Appl Software, 2012, 29 (3): 91 (张永, 李卓然, 刘小丹. 基于主动学习 SMOTE 的非均衡数 据分类. 计算机应用与软件, 2012, 29(3): 91) [7] Wang C X, Zhang T, Ma C S. Improved SMOTE algorithm for im鄄 balanced datasets. J Frontiers Comput Sci Technol, 2014, 8(6): 727 (王超学, 张涛, 马春森. 面向不平衡数据集的改进型 SMOTE 算法. 计算机科学与探索, 2014, 8(6): 727) [8] Li J, Jin H, Liu W. Novel improved SMOTE resampling integrat鄄 ed algorithm based on fractal for geochemical anomalies evalua鄄 tion. Appl Res Comput, 2012, 29(10): 3744 (李江, 金辉, 刘伟. 基于分形 SMOTE 重采样集成算法圈定 区域化探异常. 计算机应用研究, 2012, 29(10): 3744) [9] Liu X Y, Wu J X, Zhou Z H. Exploratory under鄄sampling for class鄄imbalance learning. IEEE Trans Syst Man Cybernetics Part B Cybernetics, 2009, 39(2): 539 [10] Mani I, Zhang I. kNN approach to unbalanced data distribu鄄 tions: a case study involving information extraction / / Proceed鄄 ings of the ICML 2003 Workshop on Learning from Imbalanced Datasets. Washington DC,2003: 42 [11] Kubat M, Matwin S. Addressing the curse of imbalanced training sets: one鄄sided selection / / International Conference on Machine Learning. Scotland, 2012: 179 [12] Zhu Y Q, Deng W B. A method using clustering and sampling approach for imbalance data. J Nanjing Univ Nat Sci, 2015 ,51 (2): 421 (朱亚奇, 邓维斌. 一种基于不平衡数据的聚类抽样方法. 南京大学学报(自然科学版), 2015, 51(2): 421) [13] Dietterich T G. Machine learning research: four current direc鄄 tions. Artif Intell Mag, 1997, 18(4): 97 [14] Harrington P. Translated by Qu Y D, Li R, Wang B, et al. Ma鄄 chine Learning in Action. Beijing: People蒺s Posts and Telecom鄄 munications Press, 2013 (Peter Harrington. 曲亚东, 李锐, 王斌, 等译. 机器学习实 战. 北京: 人民邮电出版社, 2013) [15] Li Y J, Guo H X, Li Y N, et al. A boosting based ensemble learning algorithm in imbalanced data classification. Syst Eng Theory Practice, 2016, 36(1): 189 (李诒靖, 郭海湘, 李亚楠, 等. 一种基于 Boosting 的集成学 习算法在不均衡数据中的分类. 系统工程理论与实践, 2016, 36(1): 189) ·1253·