Table 1 Summary of recent literature on solving data imbalance problems
From: Research on imbalance machine learning methods for MR\(T_1\)WI soft tissue sarcoma data
Ref | Year | Dataset | Methods | Evaluation metric |
|---|---|---|---|---|
[27] | 2011 | National Inpatient Sample (NIS) data | Repeated random subsampling-RF | AUC = 88.79% |
[28] | 2014 | Real datasets of human protein | MTD-SVM | AC = 96.71% |
[29] | 2021 | From Hospital Israelita Albert Einstein | MiDT | AC = 93.255% |
[30] | 2022 | The esophageal cancer patient dataset | GDO-SVM | AUC = 0.71 |
[30] | 2022 | Wisconsin | GDO-SVM | AUC = 0.9662 |
[31] | 2020 | HTRU2 | Hybrid resampling-ETC | AC = 99.3% |
[32] | 2021 | The comments on social media platforms | RVVC-SMOTE | AC = 97% |
[33] | 2021 | UCI(fraud detection) | RONS/ROS/ROA-LR/SVM | Gmean = 0.905 |
[34] | 2021 | WCE images | BIR-CNN | AC = 99.3% |
[35] | 2021 | Chest X-ray image dataset | CNNs | AC = 99.5% |