Kernel-Based Enhanced Oversampling Method for Imbalanced Classification

Li Wenjie, Hanlin WANG, Sibo Zhu, Zhijian Li
Proceedings of the 17th Asian Conference on Machine Learning, PMLR 304:862-877, 2025.

Abstract

This paper introduces a novel oversampling technique designed to improve classification performance on imbalanced datasets. The proposed method, Kernel-Weighted SMOTE (KWSMOTE), enhances the traditional SMOTE algorithm by employing a kernel-based weighting scheme to prioritize closer neighbors, which guides a convex combination that ensures the generated samples are geometrically bounded. This dual-mechanism approach generates synthetic samples that better represent the minority class. Through experiments on multiple real-world datasets, we demonstrate that KWSMOTE outperforms existing methods in terms of F1-score, G-mean, and AUC, providing a robust solution for handling imbalanced datasets in classification tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v304-wenjie25a, title = {Kernel-Based Enhanced Oversampling Method for Imbalanced Classification}, author = {Wenjie, Li and WANG, Hanlin and Zhu, Sibo and Li, Zhijian}, booktitle = {Proceedings of the 17th Asian Conference on Machine Learning}, pages = {862--877}, year = {2025}, editor = {Lee, Hung-yi and Liu, Tongliang}, volume = {304}, series = {Proceedings of Machine Learning Research}, month = {09--12 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v304/main/assets/wenjie25a/wenjie25a.pdf}, url = {https://proceedings.mlr.press/v304/wenjie25a.html}, abstract = {This paper introduces a novel oversampling technique designed to improve classification performance on imbalanced datasets. The proposed method, Kernel-Weighted SMOTE (KWSMOTE), enhances the traditional SMOTE algorithm by employing a kernel-based weighting scheme to prioritize closer neighbors, which guides a convex combination that ensures the generated samples are geometrically bounded. This dual-mechanism approach generates synthetic samples that better represent the minority class. Through experiments on multiple real-world datasets, we demonstrate that KWSMOTE outperforms existing methods in terms of F1-score, G-mean, and AUC, providing a robust solution for handling imbalanced datasets in classification tasks.} }
Endnote
%0 Conference Paper %T Kernel-Based Enhanced Oversampling Method for Imbalanced Classification %A Li Wenjie %A Hanlin WANG %A Sibo Zhu %A Zhijian Li %B Proceedings of the 17th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Hung-yi Lee %E Tongliang Liu %F pmlr-v304-wenjie25a %I PMLR %P 862--877 %U https://proceedings.mlr.press/v304/wenjie25a.html %V 304 %X This paper introduces a novel oversampling technique designed to improve classification performance on imbalanced datasets. The proposed method, Kernel-Weighted SMOTE (KWSMOTE), enhances the traditional SMOTE algorithm by employing a kernel-based weighting scheme to prioritize closer neighbors, which guides a convex combination that ensures the generated samples are geometrically bounded. This dual-mechanism approach generates synthetic samples that better represent the minority class. Through experiments on multiple real-world datasets, we demonstrate that KWSMOTE outperforms existing methods in terms of F1-score, G-mean, and AUC, providing a robust solution for handling imbalanced datasets in classification tasks.
APA
Wenjie, L., WANG, H., Zhu, S. & Li, Z.. (2025). Kernel-Based Enhanced Oversampling Method for Imbalanced Classification. Proceedings of the 17th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 304:862-877 Available from https://proceedings.mlr.press/v304/wenjie25a.html.

Related Material