Water Quality Monitoring Using Machine Learning Model

Abstract

Water is essential for human survival, and determining its quality is crucial. Traditional laboratory testing methods, though accurate, are time-consuming and impractical for real-time monitoring. This paper introduces a real-time water quality monitoring system leveraging Internet of Things (IoT) technology. The system continuously measures key water quality parameters, including temperature, turbidity, pH, electrical conductivity, and dissolved oxygen. These parameters are evaluated using a machine learning model developed with a decision tree algorithm. The collected data is transmitted to a cloud-based dashboard for real-time monitoring. The machine learning model was trained on historical laboratory data. This model is validated by passing real-time sensor readings, this model accurately determines water quality, identifying whether the water is safe for drinking or contaminated. The system is capable of issuing immediate alerts in the event of contamination. Furthermore, the model is periodically retrained with new data to improve its accuracy and adaptability. The decision tree-based machine learning model achieves an average accuracy of 97.17% in detecting water quality.