Table 5 Performance comparison on Varanasi dataset.

From: A meta-learning ensemble framework for robust and interpretable prediction of emergency medical services demand

Metric

Variant

MLP

SVR

RF

XGB

AHELM

TBLSSVR

MHKLDMR

EM-LR

MAE

T

2.9232

3.3967

2.6476

2.5965

2.8784

3.9126

2.7891

2.5376

T+W

2.8127

3.7638

2.6929

2.5748

3.1191

6.7783

3.6611

2.5508

T+W+FS

2.7054

3.6630

2.6698

2.5748

2.8056

3.7242

2.8279

2.5784

RMSE

T

3.9277

4.4677

3.6117

3.5491

3.6228

5.0845

3.7433

3.4386

T+W

3.8367

4.8692

3.6804

3.5251

4.0324

7.8700

4.7504

3.4162

T+W+FS

3.6013

4.7013

3.6379

3.5251

3.7975

4.8589

3.7381

3.4087

MAPE

T

0.8064

3.1183

0.5715

0.5309

0.5067

0.6658

0.5443

0.4757

T+W

0.6800

4.8768

0.5994

0.5242

0.5966

0.8046

0.5257

0.4728

T+W+FS

0.5694

3.1449

0.5809

0.5242

4.1941

5.0274

4.2103

0.4775

MBE

T

– 1.7024

– 1.9599

– 1.2159

– 0.9864

– 0.9771

– 2.8415

– 1.1812

– 0.4400

T+W

– 1.3962

–2.7026

– 1.4105

– 0.9610

0.3580

– 6.7566

– 1.9198

– 0.4340

T+W+FS

– 0.8426

– 2.7866

– 1.2752

– 0.9610

– 0.3106

– 3.0897

– 0.5785

-0.2933