Paper Key : IRJ************535
Author: Minal Goraksh Tattu,Mahima Nanasaheb Nawale
Date Published: 11 Nov 2024
Abstract
Water quality is critical to protecting public health and environmental sustainability. Monitoring water quality requires frequent sampling and testing, which can be time-consuming and expensive. Therefore, an effective forecasting system is needed to help ensure that the water management process is accurate and fast. Static water quality monitoring systems can be limited by human intervention and insufficient data collection. Machine learning algorithms provide an effective alternative to real-time water forecasting by processing big data. In this study, we use different types of machine learning including decision trees, random forests, gradient boosting, and neural networks to generate predictions for water parameters such as pH, turbidity, and dissolved oxygen. The data used for the training model comes from publicly available environmental databases and real-time sensor data. Each model is tuned with hyperparameters to balance prediction accuracy and computational efficiency. Our results show that random forest and gradient boosting models outperform other models and predict water quality with 92% accuracy. The results of this study demonstrate the potential of machine learning techniques in improving water quality management.
DOI Requested