Analyzing the Performance of Machine Learning Algorithms for Predicting Water Quality Index

Abstract

Abstract The aim of our research is to calculate the Water Quality Index of bore water in our surrounding educational institutions using three learning algorithms. Our research work differentiates from other work by choosing Decision Tree, K-Nearest Neighbor, and Naive Bayes and analyzing their performance with accuracy. We collected water samples from various resources and calculated the six important factors: salinity, total suspended solids (TDS), dissolved oxygen (DO), acidity and alkalinity (pH), and biochemical oxygen demand (BOD). Using efficient chemical methods, the quality parameters of water were examined. We created our dataset by utilizing these metrics, and the dataset is given as our chosen algorithm’s training and testing data. We implemented these machine learning algorithms using Google Colab. Finally, we got the WQI value with three different accuracies.