<p>Groundwater quality is a critical determinant of human health and environmental sustainability. Increasing contamination from industrial, agricultural, and urban activities necessitates the development of robust monitoring and management strategies to address this issue. This study evaluates groundwater quality in Jharkhand, India, utilizing an integrated framework that combines the Water Quality Index (WQI) and machine learning (ML) methodologies to enhance water quality classification and inform management decisions. The 2023 Jharkhand water quality dataset, encompassing locational and physicochemical parameters, was analyzed to classify water into four categories: Excellent, Good, Poor, and Unsuitable. A Random Forest Classifier, optimized via Grid Search with cross-validation and normalization, was employed to predict these classifications, achieving a performance accuracy of 90%. This approach enabled the identification of both safe and contaminated districts, providing actionable insights for public health protection, water resource management, and policy planning. Limitations include the small dataset size and reliance on a single ML algorithm, which may affect model robustness. Future work will explore larger datasets and advanced transformer-based deep learning ensembles to enhance predictive accuracy and model generalizability. The findings underscore the potential of integrating traditional water quality assessment with artificial intelligence techniques to support sustainable groundwater management and strategic decision-making.</p>