Table 8.
Difference of accuracy (%), F1-score, AUC, and AP between original and synthetic data: average over 5 ML models and five datasets with different privacy budgets ϵ = 1 and ϵ = 100.
| Method | ϵ = 1 | ϵ = 100 | ||||
|---|---|---|---|---|---|---|
| Accuracy (%) | F1-score | AUC | Accuracy (%) | F1-score | AUC | |
| CTAB-GAN+ | 32.21 | 0.473 | 0.409 | 22.04 | 0.392 | 0.348 |
| PATE-GAN | 38.17 | 0.513 | 0.447 | 37.96 | 0.508 | 0.394 |
| DP-WGAN | 36.88 | 0.536 | 0.510 | 27.27 | 0.502 | 0.423 |
| GS-WGAN | 64.10 | 0.668 | 0.495 | 58.22 | 0.639 | 0.488 |
A lower value indicates a better result. Best result in each column is highlighted in bold.