Application of differential privacy to sensor data in water quality monitoring task

Although differential privacy (DP) is used to obfuscate local information and avoid data leakage, very little research exists on the neural network model performance with applied DP for datasets from online drinking water sensor monitoring. This study aims to examine the accuracy of four different n...

Full description

Saved in:
Bibliographic Details
Main Authors: Audris Arzovs, Sergei Parshutin, Valts Urbanovics, Janis Rubulis, Sandis Dejus
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:Ecological Informatics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1574954125000287
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832576482106408960
author Audris Arzovs
Sergei Parshutin
Valts Urbanovics
Janis Rubulis
Sandis Dejus
author_facet Audris Arzovs
Sergei Parshutin
Valts Urbanovics
Janis Rubulis
Sandis Dejus
author_sort Audris Arzovs
collection DOAJ
description Although differential privacy (DP) is used to obfuscate local information and avoid data leakage, very little research exists on the neural network model performance with applied DP for datasets from online drinking water sensor monitoring. This study aims to examine the accuracy of four different neural network model architectures with DP applications. To compare the performance of the neural network model performance in total 2 215 906 augmented and experimentally obtained sensor readings were obtained from a drinking-water pilot system. Three types of contaminations at three different concentrations were applied as scenarios for anomalies in drinking water monitoring. The results achieved similar accuracy with all model architectures, with the best result showing only a 0.3% reduction in model accuracy compared with a nonprivate neural network model with 94% and 94.7% accuracy, respectively. Thus, differential privacy can be applied in the field of water quality monitoring with a reasonable decrease in the model performance.
format Article
id doaj-art-71d635048bc546abaa288936075df593
institution Kabale University
issn 1574-9541
language English
publishDate 2025-05-01
publisher Elsevier
record_format Article
series Ecological Informatics
spelling doaj-art-71d635048bc546abaa288936075df5932025-01-31T05:10:57ZengElsevierEcological Informatics1574-95412025-05-0186103019Application of differential privacy to sensor data in water quality monitoring taskAudris Arzovs0Sergei Parshutin1Valts Urbanovics2Janis Rubulis3Sandis Dejus4Institute of Electronics and Computer Science, Riga, Latvia; Corresponding authors.Institute of Information Technology, Riga Technical University, Riga, LatviaWater Systems and Biotechnology institute, Riga Technical University, Riga, LatviaWater Systems and Biotechnology institute, Riga Technical University, Riga, LatviaWater Systems and Biotechnology institute, Riga Technical University, Riga, Latvia; Corresponding authors.Although differential privacy (DP) is used to obfuscate local information and avoid data leakage, very little research exists on the neural network model performance with applied DP for datasets from online drinking water sensor monitoring. This study aims to examine the accuracy of four different neural network model architectures with DP applications. To compare the performance of the neural network model performance in total 2 215 906 augmented and experimentally obtained sensor readings were obtained from a drinking-water pilot system. Three types of contaminations at three different concentrations were applied as scenarios for anomalies in drinking water monitoring. The results achieved similar accuracy with all model architectures, with the best result showing only a 0.3% reduction in model accuracy compared with a nonprivate neural network model with 94% and 94.7% accuracy, respectively. Thus, differential privacy can be applied in the field of water quality monitoring with a reasonable decrease in the model performance.http://www.sciencedirect.com/science/article/pii/S1574954125000287Water quality monitoringDifferential privacyFederated learning
spellingShingle Audris Arzovs
Sergei Parshutin
Valts Urbanovics
Janis Rubulis
Sandis Dejus
Application of differential privacy to sensor data in water quality monitoring task
Ecological Informatics
Water quality monitoring
Differential privacy
Federated learning
title Application of differential privacy to sensor data in water quality monitoring task
title_full Application of differential privacy to sensor data in water quality monitoring task
title_fullStr Application of differential privacy to sensor data in water quality monitoring task
title_full_unstemmed Application of differential privacy to sensor data in water quality monitoring task
title_short Application of differential privacy to sensor data in water quality monitoring task
title_sort application of differential privacy to sensor data in water quality monitoring task
topic Water quality monitoring
Differential privacy
Federated learning
url http://www.sciencedirect.com/science/article/pii/S1574954125000287
work_keys_str_mv AT audrisarzovs applicationofdifferentialprivacytosensordatainwaterqualitymonitoringtask
AT sergeiparshutin applicationofdifferentialprivacytosensordatainwaterqualitymonitoringtask
AT valtsurbanovics applicationofdifferentialprivacytosensordatainwaterqualitymonitoringtask
AT janisrubulis applicationofdifferentialprivacytosensordatainwaterqualitymonitoringtask
AT sandisdejus applicationofdifferentialprivacytosensordatainwaterqualitymonitoringtask