An Oversampling Method for Class Imbalance Problems on Large Datasets
Source abstract
Abstract
Several oversampling methods have been proposed for solving the class imbalance problem. However, most of them require searching the k-nearest neighbors to generate synthetic objects. This requirement makes them time-consuming and therefore unsuitable for large datasets. In this paper, an oversampling method for large class imbalance problems that do not require the k-nearest neighbors’ search is proposed. According to our experiments on large datasets with different sizes of imbalance, the proposed method is at least twice as fast as 8 the fastest method reported in the literature while obtaining similar oversampling quality.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Imbalanced Data Classification Techniques · Anomaly Detection Techniques and Applications · Electricity Theft Detection Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Instituto Nacional de Astrofísica, Óptica y Electrónica, Tonantzintla, Puebla, Mexico City C.P. 72840, Mexico
References (62)
Show source-provided references
1. A benchmark of machine learning approaches for credit score prediction10.1016/j.eswa.2020.113986 ↗
2. Learning from class-imbalance and heterogeneous data for 30-day hospital readmission10.1016/j.neucom.2020.08.064 ↗
3. CSSG: A cost-sensitive stacked generalization approach for software defect prediction10.1002/stvr.1761 ↗
4. Sentimental analysis from imbalanced code-mixed data using machine learning approaches
5. Handling Imbalance Classification Virtual Screening Big Data Using Machine Learning Algorithms10.1155/2021/6675279 ↗
6. Vilalta, R., and Rish, I. (2003). A decomposition of classes via clustering to explain and improve naive bayes. European Conference on Machine Learning, Springer.10.1007/978-3-540-39857-8_40 ↗
7. A fine-grained random forests using class decomposition: An application to medical diagnosis10.1007/s00521-015-2064-z ↗
8. SMOTE: Synthetic minority over-sampling technique10.1613/jair.953 ↗
9. Han, H., Wang, W.Y., and Mao, B.H. (2005). Borderline-SMOTE: A new over-sampling method in imbalanced data sets learning. International Conference on Intelligent Computing, Springer.10.1007/11538059_91 ↗
10. Hu, S., Liang, Y., Ma, L., and He, Y. (2009, January 28–30). MSMOTE: Improving classification performance when training data is imbalanced. Proceedings of the 2009 Second International Workshop on Computer Science and Engineering, Qingdao, China.10.1109/WCSE.2009.756 ↗
11. Bunkhumpornpat, C., Sinapiromsaran, K., and Lursinsap, C. (2009). Safe-level-smote: Safe-level-synthetic minority over-sampling technique for handling the class imbalanced problem. Pacific-Asia Conference on Knowledge Discovery and data Mining, Springer.10.1007/978-3-642-01307-2_43 ↗
12. Maciejewski, T., and Stefanowski, J. (2011, January 11–15). Local neighbourhood extension of SMOTE for mining imbalanced data. Proceedings of the 2011 IEEE Symposium on Computational Intelligence and Data Mining (CIDM), Paris, France.10.1109/CIDM.2011.5949434 ↗
13. SMOTE-RSB*: A hybrid preprocessing approach based on oversampling and undersampling for high imbalanced data-sets using SMOTE and rough sets theory10.1007/s10115-011-0465-6 ↗
14. To Combat Multi-Class Imbalanced Problems by Means of Over-Sampling Techniques10.1109/TKDE.2015.2458858 ↗
15. Torres, F.R., Carrasco-Ochoa, J.A., and Martínez-Trinidad, J.F. (2016). SMOTE-D a deterministic version of SMOTE. Mexican Conference on Pattern Recognition, Springer.10.1007/978-3-319-39393-3_18 ↗
16. Borowska, K., and Stepaniuk, J. (2016). Imbalanced data classification: A novel re-sampling approach combining versatile improved SMOTE and rough sets. IFIP International Conference on Computer Information Systems and Industrial Management, Springer.10.1007/978-3-319-45378-1_4 ↗
17. A Novel SMOTE-Based Classification Approach to Online Data Imbalance Problem10.1155/2016/5685970 ↗
18. A novel algorithm for imbalance data classification based on genetic algorithm improved SMOTE10.1007/s13369-016-2179-2 ↗
19. Yun, J., Ha, J., and Lee, J.S. (2016, January 4–6). Automatic determination of neighborhood size in SMOTE. Proceedings of the 10th International Conference on Ubiquitous Information Management and Communication, Danang, Vietnam.10.1145/2857546.2857648 ↗
20. Oversampling the minority class in the feature space10.1109/TNNLS.2015.2461436 ↗
21. A priori synthetic over-sampling methods for increasing classification sensitivity in imbalanced data sets10.1016/j.eswa.2016.09.010 ↗
22. PSO-based method for SVM classification on skewed data sets10.1016/j.neucom.2016.10.041 ↗
23. Self-Organizing Map Oversampling (SOMO) for imbalanced data set learning10.1016/j.eswa.2017.03.073 ↗
24. Improving imbalanced learning through a heuristic oversampling method based on k-means and SMOTE10.1016/j.ins.2018.06.056 ↗
25. Adaptive semi-unsupervised weighted oversampling (A-SUWO) for imbalanced datasets10.1016/j.eswa.2015.10.031 ↗
26. Adaptive multi-objective swarm fusion for imbalanced data classification10.1016/j.inffus.2017.03.007 ↗
27. Gosain, A., and Sardana, S. (2019). Farthest SMOTE: A Modified SMOTE Approach. Computational Intelligence in Data Mining, Springer.10.1007/978-981-10-8055-5_28 ↗
28. Douzas, G., and Bacao, F. (2017). Geometric SMOTE: Effective oversampling for imbalanced learning through a geometric extension of SMOTE. arXiv.
29. An alternative SMOTE oversampling strategy for high-dimensional datasets10.1016/j.asoc.2018.12.024 ↗
30. CDSMOTE: Class decomposition and synthetic minority class oversampling technique for imbalanced-data classification10.1007/s00521-020-05130-z ↗
31. An Improved MAHAKIL Oversampling Method for Imbalanced Dataset Classification10.1109/ACCESS.2020.3047741 ↗
32. DEBOHID: A differential evolution based oversampling approach for highly imbalanced datasets10.1016/j.eswa.2020.114482 ↗
33. On oversampling imbalanced data with deep conditional generative models10.1016/j.eswa.2020.114463 ↗
34. Sadhukhan, P. (2019, January 14–19). Learning minority class prior to minority oversampling. Proceedings of the 2019 International Joint Conference on Neural Networks (IJCNN), Budapest, Hungary.10.1109/IJCNN.2019.8852188 ↗
35. Zhang, H., and Wang, Z. (2011). A normal distribution-based over-sampling approach to imbalanced data classification. International Conference on Advanced Data Mining and Applications, Springer.10.1007/978-3-642-25853-4_7 ↗
36. Rashu, R.I., Haq, N., and Rahman, R.M. (2014, January 22–23). Data mining approaches to predict final grade by overcoming class imbalance problem. Proceedings of the 2014 17th International Conference on Computer and Information Technology (ICCIT), Dhaka, Bangladesh.10.1109/ICCITechn.2014.7073095 ↗
37. Mahakil: Diversity based oversampling approach to alleviate the class imbalance issue in software defect prediction10.1109/TSE.2017.2731766 ↗
38. Tackling class imbalance problem in software defect prediction through cluster-based over-sampling with filtering10.1109/ACCESS.2019.2945858 ↗
39. An ensemble oversampling model for class imbalance problem in software defect prediction10.1109/ACCESS.2018.2817572 ↗
40. Surrounding neighborhood-based SMOTE for learning from imbalanced data sets10.1007/s13748-012-0027-5 ↗
41. Ma, L., and Fan, S. (2017). CURE-SMOTE algorithm and hybrid algorithm for feature selection and parameter optimization based on random forests. BMC Bioinform., 18.10.1186/s12859-017-1578-z ↗
42. Improved SMOTE algorithm to deal with imbalanced activity classes in smart homes10.1007/s11063-018-9940-3 ↗
43. LR-SMOTE—An improved unbalanced data set oversampling based on K-means and SVM10.1016/j.knosys.2020.105845 ↗
44. RSMOTE: A self-adaptive robust SMOTE for imbalanced problems with label noise10.1016/j.ins.2020.10.013 ↗
45. de Carvalho, A.M., and Prati, R.C. (2020). DTO-SMOTE: Delaunay Tessellation Oversampling for Imbalanced Data Sets. Information, 11.10.3390/info11120557 ↗
46. NI-MWMOTE: An improving noise-immunity majority weighted minority oversampling technique for imbalanced classification problems10.1016/j.eswa.2020.113504 ↗
47. AGNES-SMOTE: An Oversampling Algorithm Based on Hierarchical Clustering and Improved SMOTE
48. FG-SMOTE: Fuzzy-based Gaussian synthetic minority oversampling with deep belief networks classifier for skewed class distribution10.1108/IJICC-12-2020-0202 ↗
49. Mukherjee, M., and Khushi, M. (2021). SMOTE-ENC: A novel SMOTE-based method to generate synthetic data for nominal and continuous features. Appl. Syst. Innov., 4.10.3390/asi4010018 ↗
50. Loras: An oversampling approach for imbalanced datasets10.1007/s10994-020-05913-4 ↗
51. SMOTE-WENN: Solving class imbalance and small sample problems by oversampling and distance scaling10.1007/s10489-020-01852-8 ↗
52. The relationship between the mean, median, and mode with grouped data10.1080/03610926.2015.1081948 ↗
53. A parallel oversampling algorithm based on NRSBoundary-SMOTE10.12733/jics20104484 ↗
54. Comparative evaluation of pattern recognition techniques for detection of microcalcifications in mammography10.1142/S0218001493000698 ↗
55. Keel data-mining software tool: Data set repository, integration of algorithms and experimental analysis framework
56. Addressing data complexity for imbalanced data sets: Analysis of SMOTE-based oversampling and evolutionary undersampling10.1007/s00500-010-0625-8 ↗
57. Dua, D., and Graff, C. (2022, March 20). UCI Machine Learning Repository. University of California, Irvine, School of Information and Computer Sciences. Available online: http://archive.ics.uci.edu/ml.
58. Bhagat, R.C., and Patil, S.S. (2015, January 12–13). Enhanced SMOTE algorithm for classification of imbalanced big-data using random forest. Proceedings of the 2015 IEEE International Advance Computing Conference (IACC), Banglore, India.10.1109/IADCC.2015.7154739 ↗
59. SMOTE-GPU: Big data preprocessing on commodity hardware for imbalanced classification10.1007/s13748-017-0128-2 ↗
60. An insight into imbalanced big data classification: Outcomes and challenges10.1007/s40747-017-0037-9 ↗
61. A survey on addressing high-class imbalance in big data10.1186/s40537-018-0151-6 ↗
62. Basgall, M.J., Hasperué, W., Naiouf, M., Fernández, A., and Herrera, F. (2018). SMOTE-BD: An Exact and Scalable Oversampling Method for Imbalanced Classification in Big Data. VI Jornadas de Cloud Computing & Big Data (JCC & BD) (La Plata, Argentina, 2018), Universidad Nacional de La Plata.10.24215/16666038.18.e23 ↗