Water Quality Assessment using Ensemble Learning: Comparative Analysis of Stacking Classifiers for Agricultural Suitability

This research explores the use of stacking classifiers with meta-learners to classify water quality for agricultural applications. By leveraging machine learning models such as Logistic Regression, Extra Trees Classifier, K-Nearest Neighbors, and Gradient Boosting Classifier, a robust framework for predicting water suitability was developed. The dataset was preprocessed and augmented to improve model performance. Among the models, the Gradient Boosting Classifier meta-learner achieved the highest test accuracy of 96.01%. The results highlight the potential of machine learning for real-time water quality monitoring, offering a scalable solution to support sustainable agriculture.