Implementation of machine learning methods for monitoring and predicting water quality parameters
The importance of good water quality for human use and consumption can never be underestimated, and its quality is determined through effective monitoring of the water quality index. Different approaches have been employed in the treatment and monitoring of water quality parameters (WQP). Presently,...
Saved in:
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
AMG Transcend Association
2023
|
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| id |
my.uniten.dspace-26613 |
|---|---|
| record_format |
dspace |
| spelling |
my.uniten.dspace-266132023-05-29T17:12:47Z Implementation of machine learning methods for monitoring and predicting water quality parameters Hayder G. Kurniawan I. Mustafa H.M. 56239664100 56541431000 57217195204 The importance of good water quality for human use and consumption can never be underestimated, and its quality is determined through effective monitoring of the water quality index. Different approaches have been employed in the treatment and monitoring of water quality parameters (WQP). Presently, water quality is carried out through laboratory experiments, which requires costly reagents, skilled labor, and consumes time. Thereby making it necessary to search for an alternative method. Recently, machine learning tools have been successfully implemented in the monitoring, estimation, and predictions of river water quality index to provide an alternative solution to the limitations of laboratory analytical methods. In this study, the potentials of one of the machine learning tools (artificial neural network) were explored in the predictions and estimation of the Kelantan River basin. Water quality data collected from the 14 stations of the River basin was used for modeling and predicting (WQP). As for WQP analysis, the results obtained from this study show that the best prediction was obtained from the prediction of pH. The low kurtosis values of pH indicate that the appearance of outliers give a negative impact on the performance. As for WQP analysis for each station, we found that the WQP prediction in station 1, 2, and 3 give the good results. This is related to the available data of those stations that are more than the available data in other stations, except station 8. � 2020 by the authors. Final 2023-05-29T09:12:47Z 2023-05-29T09:12:47Z 2021 Article 10.33263/BRIAC112.92859295 2-s2.0-85091132978 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85091132978&doi=10.33263%2fBRIAC112.92859295&partnerID=40&md5=a21dcd384728ca33db328e207d3476f1 https://irepository.uniten.edu.my/handle/123456789/26613 11 2 9285 9295 All Open Access, Bronze AMG Transcend Association Scopus |
| institution |
Universiti Tenaga Nasional |
| building |
UNITEN Library |
| collection |
Institutional Repository |
| continent |
Asia |
| country |
Malaysia |
| content_provider |
Universiti Tenaga Nasional |
| content_source |
UNITEN Institutional Repository |
| url_provider |
http://dspace.uniten.edu.my/ |
| description |
The importance of good water quality for human use and consumption can never be underestimated, and its quality is determined through effective monitoring of the water quality index. Different approaches have been employed in the treatment and monitoring of water quality parameters (WQP). Presently, water quality is carried out through laboratory experiments, which requires costly reagents, skilled labor, and consumes time. Thereby making it necessary to search for an alternative method. Recently, machine learning tools have been successfully implemented in the monitoring, estimation, and predictions of river water quality index to provide an alternative solution to the limitations of laboratory analytical methods. In this study, the potentials of one of the machine learning tools (artificial neural network) were explored in the predictions and estimation of the Kelantan River basin. Water quality data collected from the 14 stations of the River basin was used for modeling and predicting (WQP). As for WQP analysis, the results obtained from this study show that the best prediction was obtained from the prediction of pH. The low kurtosis values of pH indicate that the appearance of outliers give a negative impact on the performance. As for WQP analysis for each station, we found that the WQP prediction in station 1, 2, and 3 give the good results. This is related to the available data of those stations that are more than the available data in other stations, except station 8. � 2020 by the authors. |
| author2 |
56239664100 |
| author_facet |
56239664100 Hayder G. Kurniawan I. Mustafa H.M. |
| format |
Article |
| author |
Hayder G. Kurniawan I. Mustafa H.M. |
| spellingShingle |
Hayder G. Kurniawan I. Mustafa H.M. Implementation of machine learning methods for monitoring and predicting water quality parameters |
| author_sort |
Hayder G. |
| title |
Implementation of machine learning methods for monitoring and predicting water quality parameters |
| title_short |
Implementation of machine learning methods for monitoring and predicting water quality parameters |
| title_full |
Implementation of machine learning methods for monitoring and predicting water quality parameters |
| title_fullStr |
Implementation of machine learning methods for monitoring and predicting water quality parameters |
| title_full_unstemmed |
Implementation of machine learning methods for monitoring and predicting water quality parameters |
| title_sort |
implementation of machine learning methods for monitoring and predicting water quality parameters |
| publisher |
AMG Transcend Association |
| publishDate |
2023 |
| _version_ |
1806424041486024704 |
| score |
13.211869 |