Table 10 Statistic metrics obtained by W-MARS model to forecast the EC parameter for all combinations.

From: An improved adaptive neuro fuzzy inference system model using conjoined metaheuristic algorithms for electrical conductivity prediction

Model

Criteria

Combination

Combo 1

Combo 2

Combo 3

Combo 4

Wavelet-Demy Test

R

0.846

0.957

0.958

0.977

RMSE

194.178

104.777

102.849

77.944

MAE

141.963

81.975

80.046

64.589

RAE

0.512

0.295

0.288

0.233

MAPE

7.543

4.213

4.138

3.358

E

0.699

0.912

0.916

0.951

IA

0.914

0.975

0.976

0.987

PI

1.000

0.691

0.684

0.612

Wavelet-Bior Test

R

0.853

0.943

0.942

0.978

RMSE

187.507

123.239

121.978

77.937

MAE

142.228

97.601

93.782

63.880

RAE

0.512

0.352

0.338

0.230

MAPE

7.500

5.102

5.059

3.337

E

0.719

0.879

0.881

0.951

IA

0.910

0.963

0.965

0.987

PI

1.000

0.768

0.758

0.616

Wavelet-Demy Train

R

0.934

0.976

0.977

0.983

RMSE

237.157

144.794

143.518

123.956

MAE

176.953

110.228

107.859

95.326

RAE

0.327

0.203

0.199

0.176

MAPE

9.814

6.030

5.973

5.437

E

0.873

0.953

0.954

0.965

IA

0.965

0.988

0.988

0.991

PI

1.000

0.760

0.754

0.712

Wavelet-Bior Train

R

0.926

0.956

0.975

0.982

RMSE

252.313

196.290

148.253

126.166

MAE

186.700

151.103

115.537

97.569

RAE

0.344

0.279

0.213

0.180

MAPE

10.175

8.270

6.406

5.568

E

0.857

0.913

0.950

0.964

IA

0.960

0.977

0.987

0.991

PI

1.000

0.872

0.754

0.699

  1. Significant values are in bold.
Back to article page