Performance of machine learning methods in predicting water quality index based on irregular data set: application on Illizi region (Algerian southeast)

Abstract Groundwater quality appraisal is one of the most crucial tasks to ensure safe drinking water sources. Concurrently, a index (WQI) requires some parameters. Conventionally, WQI computation consumes time and often found with various errors during subindex calculation. To this end, 8 artificial intelligence algorithms, e.g., multilinear regression (MLR), random forest (RF), M5P tree (M5P), subspace (RSS), additive (AR), neural network (ANN), support vector (SVR), locally weighted linear (LWLR), were employed generate prediction in Illizi region, southeast Algeria. Using best subset regression, 12 different input combinations developed strategy work was based on two scenarios. The first scenario aims reduce consumption computation, where all parameters used as inputs. second intends show variation critical cases when necessary analyses are unavailable, whereas inputs reduced sensitivity analysis. models appraised using several statistical metrics including correlation coefficient (R), mean absolute error (MAE), root square (RMSE), relative (RAE), (RRSE). results reveal that TDS TH key drivers influencing study area. comparison performance evaluation metric shows MLR model has higher accuracy compared other terms 1, 1.4572*10–08, 2.1418*10–08, 1.2573*10–10%, 3.1708*10–08% for R, MAE, RMSE, RAE, RRSE, respectively. executed less rate by RF 0.9984, 1.9942, 3.2488, 4.693, 5.9642 outcomes paper would be interest planners improving sustainable management plans groundwater resources.