@Article{cmes.2026.077236,
AUTHOR = {Md. Mottahir Alam, Mohammed K. Al Mesfer, Haroonhaider Sidhwa, Mohd Danish, Asif Irshad Khan, Tauheed Khan Mohd},
TITLE = {An Interpretable AI Framework for Predicting Groundwater Contamination under Atmospheric and Industrial Pollution Using Metaheuristic-Optimized Deep Learning},
JOURNAL = {Computer Modeling in Engineering \& Sciences},
VOLUME = {146},
YEAR = {2026},
NUMBER = {3},
PAGES = {--},
URL = {http://www.techscience.com/CMES/v146n3/66799},
ISSN = {1526-1506},
ABSTRACT = {Ground water is a crucial ecological resource and source of drinking water to a great percentage of the world population. The quality of groundwater in an area with industrial emission and air pollution is an especially important issue that requires proper evaluation. This paper introduces a spatiotemporal deep learning model that incorporates the use of metaheuristic optimization in predicting groundwater quality in various pollution contexts. The given method is a combination of the Spatial–Temporal-Assisted Deep Belief Network (StaDBN) and a hybrid Whale Optimization Algorithm and Tiki-Taka Algorithms (WOA–TTA) that would model intricate patterns of contamination. Historical ground water data sets with the hydrochemical data and time are preprocessed and pertinent and non-redundant features are determined with the Addax Optimization Algorithm (AOA). Spatial and temporal dependencies are explicitly integrated in StaDBN architecture to facilitate representation learning, and network hyperparameters are optimized by the WOA-TTA module to increase the training efficiency and predictive performance. The model was coded in Python and tested based on common statistical measures, such as root mean square error (RMSE), Nash Sutcliffe efficiency (NSE), mean absolute error (MAE), and the correlation coefficient (R). The proposed GWQP-StaDBN-WOA-TTA framework demonstrates superior predictive performance and interpretability compared to conventional machine learning and deep learning models, achieving higher correlation (R = 0.963), improved Nash–Sutcliffe efficiency (NSE = 0.84), and substantially lower prediction errors (MAE = 0.29, RMSE = 0.48), thereby validating its effectiveness for groundwater quality assessment under industrial and atmospheric pollution scenarios.},
DOI = {10.32604/cmes.2026.077236}
}