Water Quality Classification Using Machine Learning Algorithms

Abstract

Over the last several years, many more contaminants have had a substantial impact on the quality of the water. It directly affects both the environment and human health. The WQI serves as a gauge for effective water management. The fight against water pollution benefits from knowledge about water quality, including how to model it for predictions. Establishing a trustworthy prediction model for river water quality that can determine the index value based on river water quality requirements is the study's main goal. In order to determine which characteristics were important in determining the quality of river water, this study will examine and contrast the performance of many classification models and algorithms. Eleven sample sites have been selected for the data collecting process, which are dispersed among various locations along the river that flows through Kerala and Tamil Nadu. The dissolved oxygen content, temperature, pH, hardness, chloride, and other seven environmental parameters that impact the quality of water are used to calculate the water quality index. Water quality prediction model is developed using supervised machine learning methods, such as logistic regression and support vector repressor. To classify the water quality index, a classification model was created using SVM classifiers. The SVM classifier classifies the water quality index with an accuracy of 83%, whereas the logistic repressor predicts the water quality index well. The developed models performed well in terms of categorization and prediction of the water quality index.