This study applies machine learning methods to evaluate water quality in several water bodies of Jalisco, Mexico. We selected twelve critical parameters based on the Mexican Water Quality Index (ICA), as defined by Mexico’s National Water Commission (CONAGUA). These parameters include pH, electrical conductivity, dissolved oxygen, biochemical and chemical oxygen demand, nutrients, heavy metals, and microbial indicators. We trained two supervised classification models, Decision Tree and Random Forest, and one unsupervised K-Means clustering model to classify pollution levels and identify contamination patterns. Both classifiers showed good accuracy, with Random Forest outperforming Decision Tree. The K-Means algorithm revealed spatial pollution patterns, highlighting fecal coliforms as the primary contaminant. Our findings offer a clear, reproducible framework for monitoring water quality, with practical implications for regional pollution control strategies.