Performance Analysis of Machine Learning Classification Model and Ensemble Learning for Water Quality Index of Taal Lake

Acquiring accurate and reliable algorithms for examining and controlling water quality is fundamental, given its significant influence on environmental governance. Conventional Water Quality Index (WQI) standards frequently depend on basic quantitative techniques, leading to erroneous forecasts. The target of this study is to address the restraints of typical approaches by creating a Machine Learning (ML) algorithm that can precisely forecast the WQI formed on physicochemical attributes. Additionally, the study aims to compare numerous machine-learning classification models for the WQI of Taal Lake. The proposed method successfully tackles the challenge of capturing intricate non-linear connections between physicochemical properties and water quality. In this study, a variety of different machine learning approaches—including Support Vector Machines (SVM), Decision Trees, K-Nearest Neighbors (KNN), Naive Bayes, and Ensemble Random Forest are used to perform a comparative evaluation of several models that classify water based on its quality. The classification algorithms are trained employing the estimated Weighted Arithmetic Water Quality Index (WAWQI) method. It has been concluded that the Ensemble Random Forest and DT approach demonstrated exceptional performance as a classification model, obtaining a perfect accuracy rate of 100% across four evaluation metrics. This work highlights the capability of machine learning methods to enhance Water Quality Index (WQI) models and facilitate more informed decision-making in water quality management. The suggested approach may enhance worldwide water quality organization methods by providing a more precise and dependable forecast of WQI.