What question did this study set out to answer?

The research aims to develop a decision support tool that predicts water quality and recommends bioremediation strategies using machine learning methods.

March 13, 2026

A decision support machine learning tool for environmental bioremediation as water safety in India

Key Points

The research aims to develop a decision support tool that predicts water quality and recommends bioremediation strategies using machine learning methods.
Developed a Decision Support System integrating Stacking Ensemble learning and Shapley Additive ExPlanations.
Utilized Indian datasets comprising over 4000 water samples for preparation and validation.
Implemented various classifiers including RF, XGBoost, SVM, KNN, and LR to achieve prediction accuracy.
Achieved a prediction accuracy of 74.0% and an Area under the Curve (AUC) of 0.806.
Identified significant contamination drivers, such as pH and Sulfates.
Provided recommendations for ecological treatments, including phytoremediation and microbial degradation.

Abstract

This study presents a novel, interpretable Decision Support System that integrates Stacking Ensemble learning with Shapley Additive exPlanations to predict water potability and recommend bioremediation strategies. Preparation and validation based on Indian datasets with over 4000 samples was carried out and the proposed Stacking Ensemble (RF, XGBoost, SVM, KNN, LR) achieved the highest accuracy of 74.0%, with Area under the Curve of 0.806, being at about ten percent above other classifiers, and maintaining state-of-the-art level standards. Further beyond prediction, the framework locates significant contamination drivers (such as pH and Sulfates) and maps them to viable ecological treatments such as phytoremediation and microbial degradation, which recent studies lack. This study demonstrates a scalable, transparent AI-driven pathway for real-time water quality management in resource-constrained environments.

Bookmark

Cite This Study

Debnath et al. (Wed,) studied this question.

synapsesocial.com/papers/69b3abf602a1e69014ccd44d https://doi.org/https://doi.org/10.1051/bioconf/202622801003/pdf

Bookmark