Table 2 Comparative performance of proposed model vs. State-of-the-Art techniques.
Reference | Model Type | Dataset Size | R² Score | MSE (MPa²) | SHAP Max Value | LIME Fidelity |
|---|---|---|---|---|---|---|
Jibril et al.3 | Evolutionary Algorithm-Based | 900 | 0.84 | 3.20 | 0.15 | 0.80 |
Li et al.8 | XGBoost with Squirrel Search Tuning | 1000 | 0.86 | 3.00 | 0.18 | 0.82 |
Kashem & Das39 | Hybrid ML with SHAP | 850 | 0.88 | 2.80 | 0.20 | 0.85 |
Shubham et al.48 | Deep Neural Network | 950 | 0.87 | 2.95 | 0.17 | 0.83 |
Saxena et al.15 | Regression Analysis | 800 | 0.88 | 2.90 | 0.18 | 0.84 |
Proposed Model (Current Study) | Hybrid XGBoost-DNN + AutoML + MTL | 1030 | 0.91 | 2.45 | 0.25 | 0.88 |
