The purpose of this study is to investigate and compare the efficiency of artificial neural network and random forest models in predicting the qualitative and quantitative parameters of groundwater in Dehloran County, Ilam Province.

Method: To investigate the qualitative and quantitative parameters of groundwater samples and their effects on groundwater quality, 545 deep and semi-deep wells of the monitoring network of the Ilam Province Water Company were selected during a 38-year period (1986-2024) and the measured parameters EC, SAR, pH, water temperature, and water surface depth were used. Among the environmental (auxiliary) data factors in this study are geological data, land use, vegetation cover indices, derivatives from digital elevation models, distance from mines, distance from roads, distance from rivers, distance from residential areas, rainfall, and population.

Findings: The results of modeling the parameters EC, SAR, pH, water temperature and groundwater depth using artificial neural network and random forest models indicate the appropriate accuracy of the artificial neural network model in predicting these parameters, so that the final map predicted the parameters EC, SAR, pH, water temperature and groundwater depth with coefficients of explanation of 0.95, 0.93, 0.64, 0.80 and 0.87, respectively, while the random forest model predicted the parameters EC, SAR, pH, water temperature and groundwater depth with coefficients of explanation of 0.40, 0.77, 0.41, 0.74 and 0.84, respectively.

Discussion and Conclusion: Based on the results obtained, it can be concluded that the artificial neural network model has a higher accuracy than the random forest model in predicting groundwater quality parameters. By selecting the appropriate type and number of input factors, using the appropriate and compatible type of artificial neural network, and calibrating it appropriately,