Table 6.
Mean testing performance (in [%]) of 50 repetitions of semisupervised and supervised FL models. Performance comparison of semisupervised FL models when utilizing different feature importance analysis methods. Fixed parameters: In semisupervised learning, the proportion for PU1 is 0.3, and for PU2, it is 0.5. The number of trees in SecureBoost{1, 2, 3} is 10, 20, and 30, respectively. In supervised learning, the number of trees is 30 and the depth of the trees is 3.
| Methods | Acc | UAR | UF1 |
|---|---|---|---|
| SecureBoost (Supervised) | 84.628 | 84.971 | 85.107 |
| FedPU [34](SHAP feature set) | 75.103 | 68.330 | 68.001 |
| FedMatch [35] (SHAP feature set) | 65.007 | 65.791 | 62.205 |
| SecureBoost with naive PU | - | - | - |
| gain | 81.353 | 81.383 | 81.347 |
| total_gain | 80.977 | 80.964 | 80.964 |
| cover | 79.473 | 79.600 | 79.473 |
| total_cover | 81.353 | 81.383 | 81.347 |
| weight | 80.601 | 80.743 | 80.601 |
| SHAP | 84.360 | 84.338 | 84.351 |