Data-Audited Machine Learning for Reliable Water Quality Prediction and Classification in Smallholder Aquaculture

Stable water quality is essential for aquaculture productivity, but Internet of Things (IoT) sensor records from small farms may contain faulty readings that reduce the reliability of data-driven decisions. This study developed and evaluated a data-audited machine-learning decision-support pipeline for smallholder aquaculture water-quality monitoring. The workflow combined site-level sensor screening, pH estimation, feature-importance interpretation, water-status classification, dashboard visualisation, and Telegram alerting. Historical IoT records from a tilapia farm were screened using physical-validity rules before model development. Sites 1 and 2 were retained for modelling, while Site 5 was restricted to the audit because 195 of its 211 pH readings were physically invalid. The final regression and classification datasets contained 126 and 174 samples, respectively. An 80:20 split was used for both tasks, with stratification applied to classification. The Random Forest Regressor estimated pH from temperature, turbidity, and total dissolved solids with R-squared = 0.9337, RMSE = 0.0859, and MAE = 0.0471. Mean-decrease-in-impurity importance ranked TDS first at 64.50%, followed by temperature at 20.36% and turbidity at 15.15%. The Random Forest Classifier achieved 97.14% accuracy and 94.12% Critical-condition recall, but one of 17 Critical samples was misclassified, giving a 5.88% false-negative rate. Because the class labels were generated from fixed thresholds, the rule-only baseline reproduced those labels by construction and remained the authoritative safety layer. The dashboard therefore used machine learning as supplementary decision support while retaining a deterministic override for unsafe conditions in data-limited farm settings. The findings support an audited, interpretable, and conservative deployment strategy rather than reliance on predictive accuracy alone.

Aquaculture 4.0, Data quality audit, Random forest, Sensor reliability, Water quality prediction

1. Prapti, D. R., Mohamed Shariff, A. R., Che Man, H., Mohamed Ramli, N., Perumal, T., & Shariff, M. (2022). Internet of Things (IoT)-based aquaculture: An overview of IoT application on water quality monitoring. Reviews in Aquaculture, 14(2), 979–992. https://doi.org/10.1111/raq.12637 [Google Scholar] [Crossref]

2. Rahu, M. A., Chandio, A. F., Aurangzeb, K., Karim, S., Alhussein, M., & Anwar, M. S. (2023). Toward design of Internet of Things and machine learning-enabled frameworks for analysis and prediction of water quality. IEEE Access, 11, 101055–101086. https://doi.org/10.1109/ACCESS.2023.3315649 [Google Scholar] [Crossref]

3. Najah Ahmed, A., Binti Othman, F., Abdulmohsin Afan, H., Khaleel Ibrahim, R., Ming Fai, C., Shabbir Hossain, M., Ehteram, M., & Elshafie, A. (2019). Machine learning methods for better water quality prediction. Journal of Hydrology, 578, Article 124084. https://doi.org/10.1016/j.jhydrol.2019.124084 [Google Scholar] [Crossref]

4. Bui, D. T., Khosravi, K., Tiefenbacher, J., Nguyen, H., & Kazakis, N. (2020). Improving prediction of water quality indices using novel hybrid machine-learning algorithms. Science of the Total Environment, 721, Article 137612. https://doi.org/10.1016/j.scitotenv.2020.137612 [Google Scholar] [Crossref]

5. Lap, B. Q., Phan, T.-T.-H., Nguyen, H.-D., Quang, L. X., Hang, P. T., Phi, N. Q., Hoang, V. T., Linh, P. G., & Hang, B. T. T. (2023). Predicting water quality index (WQI) by feature selection and machine learning: A case study of An Kim Hai irrigation system. Ecological Informatics, 74, Article 101991. https://doi.org/10.1016/j.ecoinf.2023.101991 [Google Scholar] [Crossref]

6. Abdul Malek, N. H., Wan Yaacob, W. F., Md Nasir, S. A., & Shaadan, N. (2022). Prediction of water quality classification of the Kelantan River Basin, Malaysia, using machine learning techniques. Water, 14(7), Article 1067. https://doi.org/10.3390/w14071067 [Google Scholar] [Crossref]

7. Danuri, D., & Pozi, M. S. M. (2024). Machine learning approaches for fish pond water quality classification: Random forest, Gaussian Naive Bayes, and decision tree comparison. In Proceedings of the 11th International Applied Business and Engineering Conference (ABEC 2023). https://doi.org/10.4108/eai.21-9-2023.2342964 [Google Scholar] [Crossref]

8. Alnemari, A. M., Elmessery, W. M., Qazaq, A. S., Moustapha, M. E., Rakhimgaliyeva, S., Abuhussein, M. F. A., Alhag, S. K., Al-Shuraym, L. A., Moghanm, F. S., Szűcs, P., Eid, M. H., & Elwakeel, A. E. (2025). Developing highly accurate machine learning models for optimizing water quality management decisions in tilapia aquaculture. Scientific Reports, 15, Article 35600. https://doi.org/10.1038/s41598-025-16939-w [Google Scholar] [Crossref]

9. Essamlali, I., Nhaila, H., & El Khaili, M. (2024). Advances in machine learning and IoT for water quality monitoring: A comprehensive review. Heliyon, 10(6), Article e27920. https://doi.org/10.1016/j.heliyon.2024.e27920 [Google Scholar] [Crossref]

10. Lokman, A., Wan Ismail, W. Z., & Ab Aziz, N. A. (2025). A review of water quality forecasting and classification using machine learning models and statistical analysis. Water, 17(15), Article 2243. https://doi.org/10.3390/w17152243 [Google Scholar] [Crossref]

11. Dritsas, E., & Trigka, M. (2023). Efficient data-driven machine learning models for water quality prediction. Computation, 11(2), Article 16. https://doi.org/10.3390/computation11020016 [Google Scholar] [Crossref]

12. Xu, J., Xu, Z., Kuang, J., Lin, C., Xiao, L., Huang, X., & Zhang, Y. (2021). An alternative to laboratory testing: Random forest-based water quality prediction framework for inland and nearshore water bodies. Water, 13(22), Article 3262. https://doi.org/10.3390/w13223262 [Google Scholar] [Crossref]

13. Firdiani, F., Mandala, S., Adiwijaya, A., & Abdullah, A. H. (2024). WaQuPs: A ROS-integrated ensemble learning model for precise water quality prediction. Applied Sciences, 14(1), Article 262. https://doi.org/10.3390/app14010262 [Google Scholar] [Crossref]

14. Hussein, E. E., Jat Baloch, M. Y., Nigar, A., Abualkhair, H. F., Aldawood, F. K., & Tageldin, E. (2023). Machine learning algorithms for predicting the water quality index. Water, 15(20), Article 3540. https://doi.org/10.3390/w15203540 [Google Scholar] [Crossref]

15. Cai, H., Zhang, C., Xu, J., Wang, F., Xiao, L., Huang, S., & Zhang, Y. (2023). Water quality prediction based on the KF-LSTM encoder-decoder network: A case study with missing data collection. Water, 15(14), Article 2542. https://doi.org/10.3390/w15142542 [Google Scholar] [Crossref]

16. Madni, H. A., Umer, M., Ishaq, A., Abuzinadah, N., Saidani, O., Alsubai, S., Hamdi, M., & Ashraf, I. (2023). Water-quality prediction based on H2O AutoML and explainable AI techniques. Water, 15(3), Article 475. https://doi.org/10.3390/w15030475 [Google Scholar] [Crossref]

17. Grbčić, L., Družeta, S., Mauša, G., Lipić, T., Vukić Lušić, D., Alvir, M., Lučin, I., Sikirica, A., Davidović, D., Travaš, V., Kalafatović, D., Pikelj, K., Fajković, H., Holjević, T., & Kranjčević, L. (2022). Coastal water quality prediction based on machine learning with feature interpretation and spatio-temporal analysis. Environmental Modelling & Software, 155, Article 105458. https://doi.org/10.1016/j.envsoft.2022.105458 [Google Scholar] [Crossref]

18. Wang, S., Peng, H., & Liang, S. (2022). Prediction of estuarine water quality using an interpretable machine learning approach. Journal of Hydrology, 605, Article 127320. https://doi.org/10.1016/j.jhydrol.2021.127320 [Google Scholar] [Crossref]

19. Miller, T., Durlik, I., Krzemińska, A., Kisiel, A., Cembrowska-Lech, D., Spychalski, I., & Tuński, T. (2023). Predictive modeling of urban lake water quality using machine learning: A 20-year study. Applied Sciences, 13(20), Article 11217. https://doi.org/10.3390/app132011217 [Google Scholar] [Crossref]

20. Hoque, J. M. Z., Ab Aziz, N. A., Alelyani, S., Mohana, M., & Hosain, M. (2022). Improving water quality index prediction using regression learning models. International Journal of Environmental Research and Public Health, 19(20), Article 13702. https://doi.org/10.3390/ijerph192013702 [Google Scholar] [Crossref]

21. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324 [Google Scholar] [Crossref]

22. Naim, S. M., Das, P., Tiang, J.-J., & Nahid, A.-A. (2025). Aquaculture water quality classification using XGBoost classifier model optimized by the honey badger algorithm with SHAP and DiCE-based explanations. Water, 17(20), Article 2993. https://doi.org/10.3390/w17202993 [Google Scholar] [Crossref]

23. Mutiara, G. A., Alfarisi, M. R., & Meisaroh, L. (2026). Smart multisensor-based machine learning for multiclass freshwater fishpond water quality degradation classification. Engineering, Technology & Applied Science Research, 16(3), 34795–34806. https://doi.org/10.48084/etasr.17840 [Google Scholar] [Crossref]

24. Hridoy, M. A. A. M., Shawkat, A. I., Bordin, C., Acharjee, M. R., Masood, A., Baki, A. O., & Al Mamun, M. A. (2025). Advanced machine learning models for accurate water quality classification and WQI prediction: Implications for aquatic disease risk management. Science of the Total Environment, 1008, Article 180965. https://doi.org/10.1016/j.scitotenv.2025.180965 [Google Scholar] [Crossref]

25. Islam, M. M., Kashem, M. A., Alyami, S. A., & Moni, M. A. (2023). Monitoring water quality metrics of ponds with IoT sensors and machine learning to predict fish species survival. Microprocessors and Microsystems, 102, Article 104930. https://doi.org/10.1016/j.micpro.2023.104930 [Google Scholar] [Crossref]

• A Machine Learning Model for Predicting the Risk of Developing Diabetes - T2DM Using Real-World Data from Kilifi, Kenya
• AI-Powered Facial Recognition Attendance System Using Deep Learning and Computer Vision
• A Comprehensive Review on Brain Tumour Segmentation Using Deep Learning Approach
• A Scalable Retrieval-Augmented Generation Pipeline for Domain-Specific Knowledge Applications
• Predictive Maintenance in Semiconductor Manufacturing Using Machine Learning on Imbalanced Dataset