Water Quality Index Estimations Using Machine Learning Algorithms: A Case Study of Yazd-Ardakan Plain, Iran

Excessive population growth and high water demands have significantly increased extractions from deep semi-deep wells in the arid regions of Iran. This has negatively affected quality different areas. The Water Quality Index (WQI) is a suitable tool to assess such impacts. study used WQI fuzzy hierarchical analysis process index (FAHP-WQI) investigate status 96 agricultural Yazd-Ardakan Plain, Calculating time-consuming, but estimating inevitable for resources management. For this purpose, three Machine Learning (ML) algorithms, namely, Gene Expression Programming (GEP), M5P Model tree, Multivariate Adaptive Regression Splines (MARS), were employed predict WQI. Using Wilcox Schoeller charts, was also investigated drinking purposes. results demonstrated that 75% 33% area good quality, based on FAHP-WQI methods, respectively. According chart, around 37.25% are C3S2 C3S1 classes, which indicate poor quality. Schoeller’s diagram placed plain acceptable, inadequate, inappropriate categories. Afterwards, WQI, predicted by means ML models, compared several statistical criteria. Finally, comparative revealed MARS slightly more accurate than model