Enhancing relative humidity modelling using L2 regularization updates

Abstract This study explores L2 regularization to mitigate overfitting in artificial neural networks (ANNs), focusing on the regularization coefficient, Lambda, and its effect on data distribution and multi-layer perceptron (MLP) performance. Meteorological data from Tangier (1985–2022) with eight v...

Full description

Saved in:
Bibliographic Details
Main Authors: Abdellah Ben Yahia, Iman Kadir, Abdelaziz Abdallaoui, Abdellah El-Hmaidi
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-94356-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850042386362662912
author Abdellah Ben Yahia
Iman Kadir
Abdelaziz Abdallaoui
Abdellah El-Hmaidi
author_facet Abdellah Ben Yahia
Iman Kadir
Abdelaziz Abdallaoui
Abdellah El-Hmaidi
author_sort Abdellah Ben Yahia
collection DOAJ
description Abstract This study explores L2 regularization to mitigate overfitting in artificial neural networks (ANNs), focusing on the regularization coefficient, Lambda, and its effect on data distribution and multi-layer perceptron (MLP) performance. Meteorological data from Tangier (1985–2022) with eight variables influencing relative humidity were analyzed using principal component analysis (PCA) and self-organizing maps (SOM). PCA identifies key correlations, such as between total precipitation and relative humidity or vapor pressure and temperature, but struggles with non-linear relationships. SOM complements PCA by highlighting data structure nuances and detect complex correlations. L2 regularization, particularly with Lambda = 0.01, effectively reduces data complexity and dispersion, preventing overfitting while enhancing prediction accuracy. Adjusting Lambda during training optimizes weight biases in Kohonen and MLP networks, improving model performance and enabling precise relative humidity prediction. This strategy demonstrates the value of combining PCA, SOM, and L2 regularization for meteorological modelling.
format Article
id doaj-art-91a651cd9dae4ff7aedc83a53fe7aa1a
institution DOAJ
issn 2045-2322
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-91a651cd9dae4ff7aedc83a53fe7aa1a2025-08-20T02:55:35ZengNature PortfolioScientific Reports2045-23222025-04-0115111410.1038/s41598-025-94356-9Enhancing relative humidity modelling using L2 regularization updatesAbdellah Ben Yahia0Iman Kadir1Abdelaziz Abdallaoui2Abdellah El-Hmaidi3Laboratory of Analytical Chemistry and Electrochemistry, Faculty of Sciences, Processes and Environment Team, URL-CNRST N 13, Moulay Ismail UniversityLaboratory of Analytical Chemistry and Electrochemistry, Faculty of Sciences, Processes and Environment Team, URL-CNRST N 13, Moulay Ismail UniversityLaboratory of Analytical Chemistry and Electrochemistry, Faculty of Sciences, Processes and Environment Team, URL-CNRST N 13, Moulay Ismail UniversityLaboratory of Analytical Chemistry and Electrochemistry, Faculty of Sciences, Processes and Environment Team, URL-CNRST N 13, Moulay Ismail UniversityAbstract This study explores L2 regularization to mitigate overfitting in artificial neural networks (ANNs), focusing on the regularization coefficient, Lambda, and its effect on data distribution and multi-layer perceptron (MLP) performance. Meteorological data from Tangier (1985–2022) with eight variables influencing relative humidity were analyzed using principal component analysis (PCA) and self-organizing maps (SOM). PCA identifies key correlations, such as between total precipitation and relative humidity or vapor pressure and temperature, but struggles with non-linear relationships. SOM complements PCA by highlighting data structure nuances and detect complex correlations. L2 regularization, particularly with Lambda = 0.01, effectively reduces data complexity and dispersion, preventing overfitting while enhancing prediction accuracy. Adjusting Lambda during training optimizes weight biases in Kohonen and MLP networks, improving model performance and enabling precise relative humidity prediction. This strategy demonstrates the value of combining PCA, SOM, and L2 regularization for meteorological modelling.https://doi.org/10.1038/s41598-025-94356-9OverfittingArtificial neural networksL2 regularizationRelative humidityAnd Tangier
spellingShingle Abdellah Ben Yahia
Iman Kadir
Abdelaziz Abdallaoui
Abdellah El-Hmaidi
Enhancing relative humidity modelling using L2 regularization updates
Scientific Reports
Overfitting
Artificial neural networks
L2 regularization
Relative humidity
And Tangier
title Enhancing relative humidity modelling using L2 regularization updates
title_full Enhancing relative humidity modelling using L2 regularization updates
title_fullStr Enhancing relative humidity modelling using L2 regularization updates
title_full_unstemmed Enhancing relative humidity modelling using L2 regularization updates
title_short Enhancing relative humidity modelling using L2 regularization updates
title_sort enhancing relative humidity modelling using l2 regularization updates
topic Overfitting
Artificial neural networks
L2 regularization
Relative humidity
And Tangier
url https://doi.org/10.1038/s41598-025-94356-9
work_keys_str_mv AT abdellahbenyahia enhancingrelativehumiditymodellingusingl2regularizationupdates
AT imankadir enhancingrelativehumiditymodellingusingl2regularizationupdates
AT abdelazizabdallaoui enhancingrelativehumiditymodellingusingl2regularizationupdates
AT abdellahelhmaidi enhancingrelativehumiditymodellingusingl2regularizationupdates