Table 12 Performance results of the top 5 directly enhanced predictive Models.

From: Comprehensive framework of machine learning and deep learning architectures with metaheuristic optimization for high-fidelity prediction of nanofluid specific heat capacity

Final Models

R2 Score

MSE

RMSE

MAE

EVS

MAPE

SMAPE

Max Error

Gradient Boosting + LR

0.99905

606.98975

24.63716

14.75201

0.99905

0.00393

0.39259

165.37062

CatBoost

0.99923

487.71781

22.08433

13.33275

0.99924

0.00351

0.35084

93.33715

XGBoost

0.99893

682.15078

26.11801

15.35519

0.99894

0.00404

0.40454

124.02148

MLP + LR

0.99927

466.06792

21.58860

13.14148

0.99927

0.00355

0.35490

104.24291

GRU

0.99901

631.57643

25.13118

16.58329

0.99915

0.00443

0.44334

114.03687

Back to article page