An Intelligent Technique for Predicting Quality of Drinking Water

Water is an essential part of our lives, but as time passes water resources are becoming polluted worldwide due to industrial pollution, human waste, and many other factors. It has many hazardous effects on humans and other living beings therefore, it became crucial to monitor water. This research aims to predict water quality based on various factors of drinking water using supervised machine learning algorithms. In the context of predicting water quality, supervised machine-learning strategies are quite important. These algorithms are trained with historical data that includes information on various factors affecting water quality. This work uses machine learning algorithms to determine drinking water quality using a public dataset. Four distinct algorithms for machine learning Random Forest, Decision Tree, K-Nearest Neighbors, and Naive Bayes are used to predict the water quality. The performance of each algorithm is evaluated and compared based on their accuracy rates. After applying algorithms, the combination of two algorithms Random Forest and Decision Tree achieved perfect accuracy of 99.68%. The outcomes of the proposed model may contribute to effective water quality monitoring.