Monitoring Water Quality Based on Potability Using Machine Learning and IoT

Around the world, water contamination is a big problem. Water quality needs to be maintained. The proposed work explores the development of a real time, cost-effective, and sustainable system for water quality monitoring. The proposed system uses various sensors to calculate several physical, as well as chemical properties of water. After examining various water sample properties, the proposed system experimented used a variety of machine learning techniques, including Support Vector Machine, Time Series, Naives Bayes, K-means, etc. In this proposed system the water sample is collected from a variety of lakes, rivers, purifiers, and rainfall systems from a particular region. Collected water samples are processed by the Arduino model then the values are measured using the sensors. After processing the water samples will be categorized into 4 properties, i.e., TDS, pH temperature, and turbidity. This sensor data is further given to machine learning models. For predictive analysis, accuracy is calculated using the sensor data that is experimented on various machine learning models based on the standard drinking water range.