Predicting total coliform levels in water accurately is important for protecting public health and the environment. In this study, Support Vector Machines (SVM) and Decision Tree models are used to predict total coliforms in Indian surface and groundwater sources using physicochemical water quality parameters as input. A dataset was prepared by referring to various research articles in the Scopus database and the same was used for training, testing, and validation (70:15:15) of the model to ensure high quality performance. Both the models were assessed using error-based metrics such as R-squared (R2), Mean Squared Error (MSE), and Mean Absolute Error (MAE). The Decision Tree Regressor outperformed SVM Regressor in predicting the total coliforms in the water samples with an R2 value of 0.704 and 0.384, respectively. The study highlights the potential of Machine learning models in predicting water quality and thereby support effective monitoring and protection of water supply sources.