Water quality assessment for aquaculture using deep neural network
Aquaculture is one of the promising sector for the economic growth of developing countries like India. Exporting of fish based foods increasing gradually that raises Gross Domestic Product (GDP) values of country. In this type of scenarios, it is very essential to take care of quality of fish ponds...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Desalination and Water Treatment |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1944398625000323 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832540394431184896 |
|---|---|
| author | Rajeshwarrao Arabelli T. Bernatin Venkataramana Veeramsetty |
| author_facet | Rajeshwarrao Arabelli T. Bernatin Venkataramana Veeramsetty |
| author_sort | Rajeshwarrao Arabelli |
| collection | DOAJ |
| description | Aquaculture is one of the promising sector for the economic growth of developing countries like India. Exporting of fish based foods increasing gradually that raises Gross Domestic Product (GDP) values of country. In this type of scenarios, it is very essential to take care of quality of fish ponds water for the growth of fish species and metabolism. In this paper, optimal DNN model is developed by tuning the various parameters like number of hidden layers, hidden neurons and activation functions using Python programme in visual studio. A new dataset is generated with 1400 excellent quality water samples, 1400 good quality water samples and 1500 poor quality water samples. This dataset consists of total 14 input parameters like temperature, turbidity, dissolved Oxygen, biochemical oxygen demand (BOD), CO2, pH, alkalinity, hardness, calcium, ammonia, nitrite, phosphorus, H2S and plankton. The proposed model is validated by comparing with other machine learning models like support vector machine, K-Nearest Neighbour and Naive Bayes classifier in terms of metrics like accuracy, f1-score, precision and recall. The proposed DNN model that has training accuracy 97.41% and testing accuracy 95.69% is used to deploy in back-end of web application to assess the quality of water in fish ponds. |
| format | Article |
| id | doaj-art-66fd4fb8e3a94cab8b4f6718b2f066b5 |
| institution | Kabale University |
| issn | 1944-3986 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Desalination and Water Treatment |
| spelling | doaj-art-66fd4fb8e3a94cab8b4f6718b2f066b52025-02-05T04:31:40ZengElsevierDesalination and Water Treatment1944-39862025-01-01321101016Water quality assessment for aquaculture using deep neural networkRajeshwarrao Arabelli0T. Bernatin1Venkataramana Veeramsetty2Department of ECE, Sathyabama Institute of Science and Technology, Chennai, Tamilnadu, India; Department of ECE, School of Engineering, SR University, Warangal, Telangana State, IndiaDepartment of ECE, Sathyabama Institute of Science and Technology, Chennai, Tamilnadu, India; Corresponding author.Center for AI and Deep Learning, School of Computer Science and AI, SR University, Warangal, Telangana State, IndiaAquaculture is one of the promising sector for the economic growth of developing countries like India. Exporting of fish based foods increasing gradually that raises Gross Domestic Product (GDP) values of country. In this type of scenarios, it is very essential to take care of quality of fish ponds water for the growth of fish species and metabolism. In this paper, optimal DNN model is developed by tuning the various parameters like number of hidden layers, hidden neurons and activation functions using Python programme in visual studio. A new dataset is generated with 1400 excellent quality water samples, 1400 good quality water samples and 1500 poor quality water samples. This dataset consists of total 14 input parameters like temperature, turbidity, dissolved Oxygen, biochemical oxygen demand (BOD), CO2, pH, alkalinity, hardness, calcium, ammonia, nitrite, phosphorus, H2S and plankton. The proposed model is validated by comparing with other machine learning models like support vector machine, K-Nearest Neighbour and Naive Bayes classifier in terms of metrics like accuracy, f1-score, precision and recall. The proposed DNN model that has training accuracy 97.41% and testing accuracy 95.69% is used to deploy in back-end of web application to assess the quality of water in fish ponds.http://www.sciencedirect.com/science/article/pii/S1944398625000323Deep neural networksAquacultureWater qualityFishDeep learning |
| spellingShingle | Rajeshwarrao Arabelli T. Bernatin Venkataramana Veeramsetty Water quality assessment for aquaculture using deep neural network Desalination and Water Treatment Deep neural networks Aquaculture Water quality Fish Deep learning |
| title | Water quality assessment for aquaculture using deep neural network |
| title_full | Water quality assessment for aquaculture using deep neural network |
| title_fullStr | Water quality assessment for aquaculture using deep neural network |
| title_full_unstemmed | Water quality assessment for aquaculture using deep neural network |
| title_short | Water quality assessment for aquaculture using deep neural network |
| title_sort | water quality assessment for aquaculture using deep neural network |
| topic | Deep neural networks Aquaculture Water quality Fish Deep learning |
| url | http://www.sciencedirect.com/science/article/pii/S1944398625000323 |
| work_keys_str_mv | AT rajeshwarraoarabelli waterqualityassessmentforaquacultureusingdeepneuralnetwork AT tbernatin waterqualityassessmentforaquacultureusingdeepneuralnetwork AT venkataramanaveeramsetty waterqualityassessmentforaquacultureusingdeepneuralnetwork |