Multi-Expression Programming (MEP): Water Quality Assessment Using Water Quality Indices

Water contamination is indeed a worldwide problem that threatens public health, environmental protection, and agricultural productivity. The distinctive attributes of machine learning (ML)-based modelling can provide in-depth understanding into increasing water quality challenges. This study presents the development multi-expression programming (MEP) based predictive model for parameters, i.e., electrical conductivity (EC) total dissolved solids (TDS) in upper Indus River at two different outlet locations using 360 readings collected on monthly basis. optimized MEP models were assessed statistical measurements coefficient-of-determination (R2), root-mean-square error (RMSE), mean-absolute (MAE), root-mean-square-logarithmic (RMSLE) mean-absolute-percent (MAPE). results show R2 testing phase (subjected to unseen data) EC-MEP TDS-MEP above 0.90, 0.9674 0.9725, respectively, reflecting higher accuracy generalized performance. Also, measures are quite lower. In accordance with MAPE statistics, both shows an “excellent” performance all three stages. comparison traditional non-linear regression (NLRMs), developed have good generalization capabilities. sensitivity analysis regard significance each input forecasted parameters suggests Cl HCO3 substantial impacts predictions (EC TDS), sensitiveness index although influence Na less prominent. this research suggest intelligence EC TDS cost effective viable evaluation monitoring river water.