Kernel-Based Enhanced Oversampling Method for Imbalanced Classification
Machine Learning
2025-04-15 v1
Abstract
This paper introduces a novel oversampling technique designed to improve classification performance on imbalanced datasets. The proposed method enhances the traditional SMOTE algorithm by incorporating convex combination and kernel-based weighting to generate synthetic samples that better represent the minority class. Through experiments on multiple real-world datasets, we demonstrate that the new technique outperforms existing methods in terms of F1-score, G-mean, and AUC, providing a robust solution for handling imbalanced datasets in classification tasks.
Cite
@article{arxiv.2504.09147,
title = {Kernel-Based Enhanced Oversampling Method for Imbalanced Classification},
author = {Wenjie Li and Sibo Zhu and Zhijian Li and Hanlin Wang},
journal= {arXiv preprint arXiv:2504.09147},
year = {2025}
}