Comparison of Self-Organizing Map, Artificial Neural Network, and Co-Active Neuro-Fuzzy Inference System Methods in Simulating Groundwater Quality: Geospatial Artificial Intelligence

Water quality experiments are difficult, costly, and time-consuming. Therefore, different modeling methods can be used as an alternative for these experiments. To achieve the research objective, geospatial artificial intelligence approaches such self-organizing map (SOM), neural network (ANN), co-active neuro-fuzzy inference system (CANFIS) were to simulate groundwater in Mazandaran plain north of Iran. Geographical information (GIS) techniques a pre-processer post-processer. Data from 85 drinking water wells was secondary data separated into two splits (a) 70 percent training (60% 10% cross-validation), (b) 30 test stage. The index (GWQI) effective factors (distance industries, depth, transmissivity aquifer formations) implemented output input variables, respectively. Statistical indices (i.e., R squared (R-sqr) mean error (MSE)) utilized compare performance three methods. results demonstrate high simulation. However, stage, CANFIS (R-sqr = 0.89) had higher than SOM 0.8) ANN 0.73) tested model estimate GWQI values on area plain. Finally, mapped GIS environment associated with manage well support contribute sustainable development goal (SDG)-6, SDG-11, SDG-13.