Identification of Water Potability Using Machine Learning Techniques

The health of the flora and fauna of the earth depends significantly on the environmental water quality. It impacts humans and other species that rely on these ecosystems for nutrient cycling. Hence, water quality is highly desirable, but no single factor can be used to measure water potability. Machine learning (ML)- and deep learning (DL)-based techniques can be trained using datasets consisting of vital factors that affect water quality to predict water potability. In this book chapter, the primary objective is to know the applicability of ML and DL techniques to predict water potability. We perform extensive experiments using real datasets where popular machine learning techniques (gradient boosting, random forest, decision tree classifier, logistic regression, k-NN, multilayer perceptron, extra trees, and naive Bayes) are applied. The results are given with proper discussions and suggestions.