第6期 石洪波，等：SMOTE过采样及其改进算法研究综述 ·1081· [19]SHI Hongbo,GAO Qigang,JI Suqin,et al.A hybrid anced learning through a heuristic oversampling method sampling method based on safe screening for imbalanced based on k-means and SMOTE[J].Information sciences, datasets with sparse structure[Cl//Proceedings of 2018 In- 2018.465:1-20. ternational Joint Conference on Neural Networks.Rio de [30]楼晓俊，孙雨轩，刘海涛.聚类边界过采样不平衡数据 Janeiro.Brazil.2018:1-8. 分类方法[J刀.浙江大学学报（工学版），2013,47(6)： [20]吴艺凡，梁吉业，王俊红.基于混合采样的非平衡数据 944950. 分类算法).计算机科学与探索，2019,132)：342-349. LOU Xiaojun,SUN Yuxuan,LIU Haitao.Clustering WU Yifan,LIANG Jiye,WANG Junhong.Classification boundary over-sampling classification method for imbal- algorithm based on hybrid sampling for unbalanced anced data sets[J].Journal of Zhejiang University (Engin- data[J].Journal of frontiers of computer science and tech- eering Science),2013,47(6):944-950 nology,2019,13(2):342-349. [31]MA Li,FAN Suohai.CURE-SMOTE algorithm and hy- [21]RAMENTOL E.CABALLERO Y.BELLO R.et al. brid algorithm for feature selection and parameter optim- SMOTE-RSB*:a hybrid preprocessing approach based on ization based on random forests[J].BMC bioinformatics, oversampling and undersampling for high imbalanced 2017,18(1):169 data-sets using SMOTE and rough sets theory[J].Know- [32]IJAZ M F,ALFIAN G,SYAFRUDIN M,et al.Hybrid ledge and information systems,2012,33(2):245-265. prediction model for type 2 diabetes and hypertension us- [22]SAEZ J A,LUENGO J,STEFANOWSKI J,et al. ing DBSCAN-based outlier detection,synthetic minority SMOTE-IPF:addressing the noisy and borderline ex- over sampling technique(SMOTE),and random forest[J]. amples problem in imbalanced classification by a re- Applied sciences,2018,8(8):1325. sampling method with filtering[J].Information sciences, [33]盛凯，刘忠，周德超，等.面向不平衡分类的IDP- 2015.291:184-203. SMOTE重采样算法).计算机应用研究，2019,36(01)： [23]RADWAN A M.Enhancing prediction on imbalance data 115-118 by thresholding technique with noise filtering[C]//Pro- SHENG Kai,LIU Zhong,ZHOU Dechao,et al.IDP- ceedings of 2017 International Conference on Informa- SMOTE resampling algorithm for imbalanced classifica- tion Technology.Amman,Jordan,2017:399-404. tion[J].Application research of computers,2019,36(01): [24]ZHANG Jianjun,NG W.Stochastic sensitivity measure- 115-118. based noise filtering and oversampling method for imbal- [34]BLAGUS R,LUSA L.SMOTE for high-dimensional anced classification problems[C]//Proceedings of 2018 class-imbalanced data[J].BMC bioinformatics,2013,14: IEEE International Conference on Systems,Man,and Cy- 106. bernetics.Miyazaki,Japan,2018:403-408. [35]ABDI L,HASHEMI S.To combat multi-class imbal- [25]BISPO A.PRUDENCIO R,VERAS D.Instance selec- anced problems by means of over-sampling techniques. tion and class balancing techniques for cross project de- IEEE transactions on knowledge and data engineering. fect prediction[Cl//Proceedings of 2018 Brazilian Confer- 2016,28(1238-251. ence on Intelligent Systems.Sao Paulo,Brazil,2018: [36]WANG Jin,YUN Bo,HUANG Pingli,et al.Applying 552-557 threshold SMOTE algorithm with attribute bagging to im- [26]BATISTA G E A P A.PRATI R C.MONARD M C.A balanced datasets[C]//Proceedings of the 8th Internation- study of the behavior of several methods for balancing al Conference on Rough Sets and Knowledge Techno- machine learning training datalJ].ACM SIGKDD explor- logy.Halifax,NS,Canada,2013:221-228. ations newsletter,2004,6(1):20-29 [37]MATHEW J,LUO Ming,PANG C K,et al.Kernel-based [27]BARUA S,ISLAM MM,YAO Xin,et al.MWMOTE- SMOTE for SVM classification of imbalanced majority weighted minority oversampling technique for datasets[C]//Proceedings of IECON 2015-41st Annual imbalanced data set learning[J].IEEE transactions on Conference of the IEEE Industrial Electronics Society knowledge and data engineering,2014,26(2):405-425. Yokohama,Japan,2015:1127-1132. [28]PRUENGKARN R.WONG K W.FUNG CC.Multi- [38]BELLINGER C,DRUMMOND C,JAPKOWICZ N class imbalanced classification using fuzzy C-mean and Beyond the boundaries of SMOTE-A framework for man- SMOTE with fuzzy support vector machine[C]//Proceed- ifold-based synthetically oversampling[C]//Proceedings of ings of the 24th International Conference on Neural In- Joint European Conference on Machine Learning and formation Processing.Guangzhou,China,2017:67-75. Knowledge Discovery in Databases.Riva del Garda, [29]DOUZAS G,BACAO F,LAST F.Improving imbal- taly,2016:248-263.SHI Hongbo, GAO Qigang, JI Suqin, et al. A hybrid sampling method based on safe screening for imbalanced datasets with sparse structure[C]//Proceedings of 2018 In￾ternational Joint Conference on Neural Networks. Rio de Janeiro, Brazil, 2018: 1−8. [19] 吴艺凡, 梁吉业, 王俊红. 基于混合采样的非平衡数据 分类算法 [J]. 计算机科学与探索, 2019, 13(2): 342–349. WU Yifan, LIANG Jiye, WANG Junhong. Classification algorithm based on hybrid sampling for unbalanced data[J]. Journal of frontiers of computer science and tech￾nology, 2019, 13(2): 342–349. [20] RAMENTOL E, CABALLERO Y, BELLO R, et al. SMOTE-RSB*: a hybrid preprocessing approach based on oversampling and undersampling for high imbalanced data-sets using SMOTE and rough sets theory[J]. Know￾ledge and information systems, 2012, 33(2): 245–265. [21] SÁEZ J A, LUENGO J, STEFANOWSKI J, et al. SMOTE–IPF: addressing the noisy and borderline ex￾amples problem in imbalanced classification by a re￾sampling method with filtering[J]. Information sciences, 2015, 291: 184–203. [22] RADWAN A M. Enhancing prediction on imbalance data by thresholding technique with noise filtering[C]//Pro￾ceedings of 2017 International Conference on Informa￾tion Technology. Amman, Jordan, 2017: 399−404. [23] ZHANG Jianjun, NG W. Stochastic sensitivity measure￾based noise filtering and oversampling method for imbal￾anced classification problems[C]//Proceedings of 2018 IEEE International Conference on Systems, Man, and Cy￾bernetics. Miyazaki, Japan, 2018: 403−408. [24] BISPO A, PRUDENCIO R, VÉRAS D. Instance selec￾tion and class balancing techniques for cross project de￾fect prediction[C]//Proceedings of 2018 Brazilian Confer￾ence on Intelligent Systems. Sao Paulo, Brazil, 2018: 552−557. [25] BATISTA G E A P A, PRATI R C, MONARD M C. A study of the behavior of several methods for balancing machine learning training data[J]. ACM SIGKDD explor￾ations newsletter, 2004, 6(1): 20–29. [26] BARUA S, ISLAM M M, YAO Xin, et al. MWMOTE￾majority weighted minority oversampling technique for imbalanced data set learning[J]. IEEE transactions on knowledge and data engineering, 2014, 26(2): 405–425. [27] PRUENGKARN R, WONG K W, FUNG C C. Multi￾class imbalanced classification using fuzzy C-mean and SMOTE with fuzzy support vector machine[C]//Proceed￾ings of the 24th International Conference on Neural In￾formation Processing. Guangzhou, China, 2017: 67−75. [28] [29] DOUZAS G, BACAO F, LAST F. Improving imbal￾anced learning through a heuristic oversampling method based on k-means and SMOTE[J]. Information sciences, 2018, 465: 1–20. 楼晓俊, 孙雨轩, 刘海涛. 聚类边界过采样不平衡数据 分类方法 [J]. 浙江大学学报 (工学版), 2013, 47(6): 944–950. LOU Xiaojun, SUN Yuxuan, LIU Haitao. Clustering boundary over-sampling classification method for imbal￾anced data sets[J]. Journal of Zhejiang University (Engin￾eering Science), 2013, 47(6): 944–950. [30] MA Li, FAN Suohai. CURE-SMOTE algorithm and hy￾brid algorithm for feature selection and parameter optim￾ization based on random forests[J]. BMC bioinformatics, 2017, 18(1): 169. [31] IJAZ M F, ALFIAN G, SYAFRUDIN M, et al. Hybrid prediction model for type 2 diabetes and hypertension us￾ing DBSCAN-based outlier detection, synthetic minority over sampling technique (SMOTE), and random forest[J]. Applied sciences, 2018, 8(8): 1325. [32] 盛凯, 刘忠, 周德超, 等. 面向不平衡分类的 IDP￾SMOTE 重采样算法 [J]. 计算机应用研究, 2019, 36(01): 115–118. SHENG Kai, LIU Zhong, ZHOU Dechao, et al. IDP￾SMOTE resampling algorithm for imbalanced classifica￾tion[J]. Application research of computers, 2019, 36(01): 115–118. [33] BLAGUS R, LUSA L. SMOTE for high-dimensional class-imbalanced data[J]. BMC bioinformatics, 2013, 14: 106. [34] ABDI L, HASHEMI S. To combat multi-class imbal￾anced problems by means of over-sampling techniques[J]. IEEE transactions on knowledge and data engineering, 2016, 28(1): 238–251. [35] WANG Jin, YUN Bo, HUANG Pingli, et al. Applying threshold SMOTE algorithm with attribute bagging to im￾balanced datasets[C]//Proceedings of the 8th Internation￾al Conference on Rough Sets and Knowledge Techno￾logy. Halifax, NS, Canada, 2013: 221–228. [36] MATHEW J, LUO Ming, PANG C K, et al. Kernel-based SMOTE for SVM classification of imbalanced datasets[C]//Proceedings of IECON 2015-41st Annual Conference of the IEEE Industrial Electronics Society. Yokohama, Japan, 2015: 1127–1132. [37] BELLINGER C, DRUMMOND C, JAPKOWICZ N. Beyond the boundaries of SMOTE-A framework for man￾ifold-based synthetically oversampling[C]//Proceedings of Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Riva del Garda, Italy, 2016: 248−263. [38] 第 6 期 石洪波，等：SMOTE 过采样及其改进算法研究综述 ·1081·