Abstract: Water quality assessment is essential for ensuring public health, environmental safety, and sustainable water resource management. Traditional methods of water monitoring rely on manual sampling and laboratory analysis, which are often time-consuming, expensive, and incapable of providing real-time insights. This study proposes a machine learning (ML)-based framework for accurate and timely prediction of water quality parameters such as pH, turbidity, dissolved oxygen, nitrates, and phosphates. Historical and sensor-based datasets are utilized to train and evaluate supervised ML models, including Random Forest (RF), Support Vector Machine (SVM), and XGBoost. Data preprocessing, feature selection, and model evaluation are incorporated to enhance prediction accuracy. Experimental results demonstrate that the proposed ML models can reliably forecast water quality metrics, providing early warnings of potential contamination events. This approach not only reduces dependence on manual testing but also supports real-time water management and pollution mitigation strategies, making it suitable for smart city and industrial applications.

Keywords: Water Quality Prediction, Machine Learning, Random Forest, XGBoost, Support Vector Machine, Environmental Monitoring, Predictive Analytics

Downloads: | DOI: 10.17148/IJIREEICE.2025.131206