Table 6 Performance evaluation of different local server client models for traffic density Prediction.
From: Enhancing smart city sustainability with explainable federated learning for vehicular energy control
Model | MSE ↓ (Lower is Better) | R2 ↑ (Higher is Better) | MAE ↓ (Lower is Better) | EVS ↑ (Higher is Better) |
|---|---|---|---|---|
MLP | 8.20e-05 (✔ Best) | 0.9983 (✔ Best) | 5.39e-03 (✔ Best) | 0.9983 (✔ Best) |
MultiOutput Decision Tree | 2.48e-08 | 0.9999 | 1.85e-06 | 0.9999 |
Decision Tree | 1.90e-03 | 0.9604 | 1.91e-02 | 0.9604 |
Elastic Net | 4.79e-02 | -7.34e-08 | 1.72e-01 | 1.57e-14 |
Orthogonal Matching Pursuit | 4.79e-02 | 8.21e-04 | 1.72e-01 | 8.21e-04 |
