From: The use of generative adversarial networks to alleviate class imbalance in tabular data: a survey
Paper - Novel vs Baseline - Sampling Method | AUC | Balanced Accuracy | F1 Score | Ranking | Recall | Specificity |
---|---|---|---|---|---|---|
Conditional Wasserstein GAN-based oversampling of tabular data for imbalanced learning | Â | Â | Â | Â | Â | Â |
 Novel |  |  |  |  |  |  |
  CWGAN |  |  |  | 3.15 |  |  |
 Baseline |  |  |  |  |  |  |
  ADAYSN |  |  |  | 5.37 |  |  |
  B-SMOTE |  |  |  | 4.03 |  |  |
  None |  |  |  | 2.32 |  |  |
  Random |  |  |  | 3.48 |  |  |
  SMOTE |  |  |  | 3.90 |  |  |
  SMOTE-ENC |  |  |  | 5.73 |  |  |
  SMOTENC |  |  |  | 5.17 |  |  |
Generative adversarial fusion network for class imbalance credit scoring | Â | Â | Â | Â | Â | Â |
 Novel |  |  |  |  |  |  |
  IGAFN | 71.08% | 83.60% | 57.98% |  |  |  |
 Baseline |  |  |  |  |  |  |
  CFN | 65.64% | 76.05% | 47.94% |  |  |  |
  GAN | 70.07% | 79.39% | 54.43% |  |  |  |
  No treatment | 63.55% | 79.35% | 42.12% |  |  |  |
  SMOTE | 65.38% | 70.98% | 43.27% |  |  |  |
Using generative adversarial networks for improving classification effectiveness in credit card fraud detection | Â | Â | Â | Â | Â | Â |
 Novel |  |  |  |  |  |  |
  GAN |  |  |  |  | 71.94% | 99.99% |
 Baseline |  |  |  |  |  |  |
  SMOTE |  |  |  |  | 70.60% | 100.00% |