Table 5 Results of one-step ahead EC forecasting for Barratta Creek.

From: Multi-step ahead forecasting of electrical conductivity in rivers by using a hybrid Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model enhanced by Boruta-XGBoost feature selection algorithm

Model

Data

R

RMSE

MAPE

E

Tstat

U95%

Boruta-XGB-CNN-LSTM

Train

0.9316

43.2172

7.6428

0.8673

2.7861

119.7122

Test

0.9215

43.8315

7.6029

0.8488

1.1701

121.4845

Boruta-XGB-MLP

Train

0.9288

44.1936

7.9495

0.8616

3.8830

122.3293

Test

0.9184

44.7175

7.7053

0.8426

2.5231

123.7993

Boruta-XGB-KNN

Train

0.9443

39.0323

7.4576

0.8918

0.1975

108.2023

Test

0.9042

48.3154

8.7854

0.8162

1.4046

133.8936

Boruta-XGB-XGBoost

Train

0.9546

35.4029

6.8040

0.9110

0.0785

98.1415

Test

0.9128

46.0644

8.5283

0.8330

0.9702

127.6856

Back to article page