Home

Computational Ecology and Software, 2025, 15(4): 157-175
[XML] [EndNote] [RefManager] [BibTex] [ Full PDF (1624K)] [Comment/Review Article]

Article

Sequential GP-UCB Bayesian optimization for deep neural network fine-tuning in dissolved oxygen prediction

Farid Hassanbaki Garabaghi1, Semra Benzer2, Recep Benzer3
1Department of Environmental Sciences, Graduate School of Natural and Applied Sciences, Gazi University 06500, Ankara, Turkey
2Department of Science Education, Gazi Faculty of Education, Gazi University 06500, Ankara, Turkey
3Department of Management Information Systems, School of Administrative and Social Sciences, Ankara Medipol University 06050, Ankara, Turkey

Received 29 March 2025;Accepted 5 May 2025;Published online 15 May 2025;Published 1 December 2025
IAEES

Abstract
Dissolved Oxygen (DO) serves as a crucial measure of water quality, imperative for both aquatic life and human consumption. The application of deep learning, particularly through data-driven predictions, offers a robust tool for estimating DO concentrations. Enhanced precision is achieved by fine-tuning hyperparameters. Bayesian optimization methods are amongst those noteworthy for their effectiveness. This study focuses on predicting DO levels using a Deep Neural Network model. The study uses Bayesian optimization to refine hyperparameters for the best model setup, comparing the results with a baseline model using default settings. Results indicate that the Bayesian-optimized model outperforms the baseline. The findings underscore the pivotal role of Bayesian optimization in elevating model performance, exhibiting robust generalization capabilities while significantly reducing the need for manual parameter tuning. This successful application underscores a substantial methodological advancement in environmental management, particularly in predictive modelling for indicators of aquatic ecosystem health.

Keywords water quality;dissolved oxygen;environmental management;deep learning;Bayesian optimisation;gaussian process.


International Academy of Ecology and Environmental Sciences. E-mail: office@iaees.org
Copyright © 2009-2025 International Academy of Ecology and Environmental Sciences. All rights reserved.
Web administrator: office@iaees.org, website@iaees.org; Last modified: 2025/6/11


Translate page to: