EN PL
PRACA ORYGINALNA
Machine Learning-Based Evaporation Prediction and Kernel Function Analysis: A Case Study of the Boukourdane Dam, Algeria
 
Więcej
Ukryj
1
Department of Civil Engineering and Hydraulic, Faculty of Sciences and Technology, University of Jijel, Algeria.
 
2
Department of hydraulic, Faculty of Technology, Badji-Mokhtar Annaba University, P.O. Box 12, 23000 Annaba, Algeria
 
3
Department of Civil Engineering, Chandigarh University, Punjab 140301, India.
 
4
University of Tusca
 
 
Data nadesłania: 09-08-2025
 
 
Data ostatniej rewizji: 06-09-2025
 
 
Data akceptacji: 16-09-2025
 
 
Data publikacji: 19-03-2026
 
 
Autor do korespondencji
Andrea Petroselli   

University of Tusca
 
 
Acta Sci. Pol. Formatio Circumiectus 2026;25(1)
 
INFORMACJE KLUCZOWE
  • Evaporation at Boukourdane Dam was estimated for better water use
  • Three machine learning models were tested with the Pearson VII kernel
  • Random Forest performed best; humidity was most influential
SŁOWA KLUCZOWE
DZIEDZINY
STRESZCZENIE
Aim of the study:
The objective of this study is to evaluate the impact of three kernel functions—Pearson VII, radial basis function (RBF), and polynomial—on the predictive performance of Support Vector Regression (SVR) and Gaussian Process Regression (GPR) models.

Material and methods:
Three machine learning models—Random Forest (RF), Support Vector Regression (SVR), and Gaussian Process Regression (GPR)—were applied, testing three kernels (Pearson VII, RBF, polynomial). The Pearson VII kernel’s novel use improves prediction under limited data. The dataset includes 240 observations over 20 years, with inputs: max/min air temperature, relative humidity, wind speed, and water temperature; output: evaporation.

Results and conclusions:
Model performance was evaluated via Correlation Coefficient (CC), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). RF outperformed GPR and SVR across kernels, achieving MAE=1.01 mm, RMSE=1.29 mm, and CC=0.81 in testing. Moreover, the Pearson VII kernel delivered the highest accuracy within both GP and SVM frameworks. Sensitivity analysis highlighted relative humidity as the most influential factor in evaporation forecasting.
ISSN:1644-0765
Journals System - logo
Scroll to top