Regularization effects on machine learning models for relative humidity prediction under semi-arid climate conditions
- 1Laboratory of Analytical Chemistry and Electrochemistry, Processes and Environment Teams, Department of Chemistry, Faculty of Sciences, Moulay Ismail University
- 2Laboratory of Water Sciences and Environmental Engineering Teams, Department of Geology, Faculty of Sciences, Moulay Ismail University
Abstract
In semi-arid regions, daily relative humidity exhibits strong nonlinear and regime-dependent variability, which limits the robustness of conventional predictive approaches. This study investigates the influence of regularization on predictive stability and generalization within two supervised learning frameworks: Support Vector Regression (SVR) and Radial Basis Function Neural Networks (RBF-NN), using long-term meteorological observations from Fez, Morocco (1985–2022). Seven atmospheric predictors representing thermodynamic, radiative, and moisture-related processes were used as inputs. For SVR, hyperparameters were optimized using Bayesian optimization with cross-validation. While RBF-NN parameters and regularization were determined using grid-search and hold-out validation. Both models were evaluated using an identical data partitioning scheme (70% training, 15% validation, 15% testing) to ensure a consistent comparison. The results indicate that regularization effects differ across the two frameworks. In SVR, the Gaussian RBF kernel achieves the highest predictive performance (R = 0.9890; MSE = 0.0016), reflecting strong representation of relative humidity dynamics under global regularization through parameter C. In contrast, the RBF-NN model exhibits higher sensitivity to regularization parameter selection due to its localized learning structure, with the optimal configuration (λ = 0.01; architecture [7–16–1]) with (R = 0.9603; MSE = 0.0141), while deviation from this range leads to overfitting or underfitting. These results demonstrate that SVR provides more stable predictive performance than RBF-NN under identical climatic conditions. The study contributes to the understanding of how model structure and regularization influence relative humidity prediction under semi-arid climatic conditions.
Related
- Data Quality Audit and Bivariate Temperature-Humidity Analysis of a Long-Term LoRaWAN Network: Anomaly Detection, PCA, Clustering and Machine Learning
- Resilient modulus characteristics and model prediction of cement phosphogypsum under dry-wet cycling conditions
- Assessing the vertical representativeness of ERA5 humidity for interpreting satellite upper-tropospheric humidity trends
- Elevated environmental humidity modulates aquaporin expression and impacts leaf hydraulic efficiency in silver birch (Betula pendula)
- Machine-learning prediction of freezing temperature in Pb-contaminated soils remediated with biochar
BibTeX
@article{kadir2026regularization,
title = {{Regularization effects on machine learning models for relative humidity prediction under semi-arid climate conditions}},
author = {Iman Kadir and Abdellah Ben yahia and Abdelaziz Abdallaoui and Abdellah El-Hmaidi},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-61638-9},
url = {https://doi.org/10.1038/s41598-026-61638-9},
note = {CC BY},
}
RIS
TY - JOUR
TI - Regularization effects on machine learning models for relative humidity prediction under semi-arid climate conditions
AU - Iman Kadir
AU - Abdellah Ben yahia
AU - Abdelaziz Abdallaoui
AU - Abdellah El-Hmaidi
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-61638-9
UR - https://doi.org/10.1038/s41598-026-61638-9
AB - In semi-arid regions, daily relative humidity exhibits strong nonlinear and regime-dependent variability, which limits the robustness of conventional predictive approaches. This study investigates the influence of regularization on predictive stability and generalization within two supervised learning frameworks: Support Vector Regression (SVR) and Radial Basis Function Neural Networks (RBF-NN), using long-term meteorological observations from Fez, Morocco (1985–2022). Seven atmospheric predictors representing thermodynamic, radiative, and moisture-related processes were used as inputs. For SVR, hyperparameters were optimized using Bayesian optimization with cross-validation. While RBF-NN parameters and regularization were determined using grid-search and hold-out validation. Both models were evaluated using an identical data partitioning scheme (70% training, 15% validation, 15% testing) to ensure a consistent comparison. The results indicate that regularization effects differ across the two frameworks. In SVR, the Gaussian RBF kernel achieves the highest predictive performance (R = 0.9890; MSE = 0.0016), reflecting strong representation of relative humidity dynamics under global regularization through parameter C. In contrast, the RBF-NN model exhibits higher sensitivity to regularization parameter selection due to its localized learning structure, with the optimal configuration (λ = 0.01; architecture [7–16–1]) with (R = 0.9603; MSE = 0.0141), while deviation from this range leads to overfitting or underfitting. These results demonstrate that SVR provides more stable predictive performance than RBF-NN under identical climatic conditions. The study contributes to the understanding of how model structure and regularization influence relative humidity prediction under semi-arid climatic conditions.
SN - 2045-2322
ER -