Table 4 Accuracy comparison across datasets with varying \(\epsilon\) and partition configurations.
Dataset | Partitions | \(\epsilon =0.3\) | \(\epsilon =0.5\) | Features | Samples | Classes | Train (%) | Test (%) |
|---|---|---|---|---|---|---|---|---|
Heart disease | 2 | 88.52 | 91.80 | 14 | 303 | 2 | 80 | 20 |
6 | 83.60 | 91.80 | ||||||
Heart failure | 2 | 86.66 | 90.00 | 13 | 299 | 2 | 80 | 20 |
6 | 73.33 | 85.00 | ||||||
Synthetic-1 | 15 | 88.86 | 95.36 | 30 | 15,000 | 17 | 80 | 20 |
21 | 75.60 | 93.73 | ||||||
Synthetic-2 | 9 | 92.82 | 94.14 | 15 | 17,000 | 7 | 80 | 20 |
27 | 87.02 | 92.70 |
