Table 3 Comparison of performance metrics across study areas and pollutants.
Model | PM2.5 | PM10 | NO2 | O3 |
|---|---|---|---|---|
RMSE (µg/m³) | ||||
MAST-Net | 8.2 ± 1.1 | 12.4 ± 1.8 | 6.7 ± 0.9 | 15.3 ± 2.1 |
CNN-LSTM | 11.6 ± 1.4 | 17.2 ± 2.3 | 9.8 ± 1.2 | 21.7 ± 2.8 |
CNN-LSTM-Attention | 10.3 ± 1.2 | 15.1 ± 2.0 | 8.4 ± 1.0 | 19.2 ± 2.4 |
Random Forest | 15.3 ± 2.1 | 22.1 ± 3.2 | 12.4 ± 1.6 | 26.8 ± 3.5 |
SVR | 18.7 ± 2.5 | 25.8 ± 3.8 | 14.2 ± 1.8 | 29.3 ± 4.1 |
Linear Regression | 22.4 ± 3.1 | 31.2 ± 4.5 | 16.8 ± 2.2 | 35.6 ± 5.2 |
R² Score | ||||
MAST-Net | 0.91 ± 0.03 | 0.89 ± 0.04 | 0.94 ± 0.02 | 0.87 ± 0.05 |
CNN-LSTM | 0.84 ± 0.04 | 0.81 ± 0.05 | 0.87 ± 0.03 | 0.79 ± 0.06 |
CNN-LSTM-Attention | 0.87 ± 0.03 | 0.84 ± 0.04 | 0.90 ± 0.02 | 0.82 ± 0.05 |
Random Forest | 0.76 ± 0.05 | 0.73 ± 0.06 | 0.79 ± 0.04 | 0.71 ± 0.07 |
MAE (µg/m³) | ||||
MAST-Net | 6.1 ± 0.8 | 9.2 ± 1.3 | 4.9 ± 0.7 | 11.8 ± 1.6 |
CNN-LSTM | 8.9 ± 1.1 | 13.5 ± 1.9 | 7.6 ± 0.9 | 17.2 ± 2.3 |
Random Forest | 12.1 ± 1.7 | 17.8 ± 2.6 | 9.8 ± 1.2 | 21.4 ± 2.9 |
