Table 4 Results of a performance evaluation using the developed ML models during training, testing and validation phase for minimum air temperature forecasting.

From: Air temperature estimation and modeling using data driven techniques based on best subset regression model in Egypt

Model

Dataset

NSE

d

MAE

RMSE

RAE

RRSE

PCC

R 2

Linear Regression

Training

0.8025

0.9424

1.952

2.503

40.506

44.441

0.8958

0.8025

Testing

0.7938

0.9397

1.902

2.453

40.528

45.208

0.8910

0.7938

Validation

0.8022

0.9424

1.953

2.505

40.528

44.470

0.8957

0.8022

Additive Regression

Training

0.7675

0.9293

2.158

2.716

44.774

48.219

0.8763

0.7678

Testing

0.7506

0.9250

2.113

2.113

45.041

49.716

0.8671

0.7519

Validation

0.7617

0.9296

2.174

2.749

45.111

48.813

0.8728

0.7619

Random Subspace

Training

0.8145

0.9436

1.916

2.42

39.765

43.067

0.9041

0.8173

Testing

0.7845

0.9350

1.962

2.507

41.809

46.213

0.8864

0.7857

Validation

0.7865

0.9344

2.051

2.602

42.546

46.202

0.8878

0.7882

M5P

Training

0.8061

0.9437

1.945

2.480

40.363

44.033

0.8978

0.8061

Testing

0.7951

0.9411

1.899

2.445

40.465

45.062

0.8919

0.7956

Validation

0.8048

0.9433

1.952

2.488

40.489

40.488

0.8971

0.8048

SVM

Training

0.8012

0.9427

1.945

2.511

40.356

44.587

0.8957

0.8022

Testing

0.7920

0.9399

1.899

2.463

40.464

45.405

0.8905

0.7930

Validation

0.8009

0.9427

1.947

2.513

40.393

44.613

0.8955

0.8020

  1. The best-performing machine learning model for forecasting daily minimum air temperature is highlighted in bold blue, while the poorest-performing model is indicated in red.
Back to article page