Water, an essential resource, has got stern importance in checking its quality due to the influence of various external factors like industrial emissions, acid rain, dumping of degradable and non-degradable wastes. Drinking impure water has a direct impact on public health and life. Hence, it is essential to ensure the quality of drinking water before consumption. This project work focuses on predicting the water quality of well water in Chengalpattu district. The work in this project is two-fold, i) classify the water quality using machine learning models such as Support Vector Machine (SVM), Decision Tree (DT) and Random Forest (RF) and ii) predict the quality using rough set theory (RST). Rough set theory is used to identify dependencies within the data and to handle uncertainty and incomplete information in water quality datasets. Rough set theory facilitates the identification of indiscernibility relations and boundary regions within the data, enabling the classification of water samples that are on the edge of different quality categories. The rough set principles are used to handle uncertainty and ambiguity in water quality classification. The water quality can be classified into four classes excellent, good, poor, and very poor. The results of this study demonstrate the efficiency of machine learning algorithms and rough set theory in water quality analysis.The models are then tested and evaluated to find the best suitable model by analysing the accuracy of prediction, the precision and recall of all the models.