Random Forest Based ML Model for Potability of Water

Access to safe drinking water is a fundamental necessity for public health, yet many regions face challenges in evaluating and ensuring water quality. We use a machine learning model to evaluate water safety by analyzing various physicochemical indicators Utilizing a publicly available dataset containing features such as pH, hardness, solids, sulfate, conductivity, organic carbon, trihalomethanes, turbidity, and others, we apply data preprocessing techniques to handle missing values and standardize feature distributions. A Random Forest classifier is then employed due to its robustness, interpretability, and ability to handle non-linear relationships. The model is trained and evaluated using a stratified train-test split, achieving strong performance in classifying water as potable or non-potable. Furthermore, the feature importance analysis highlights critical parameters influencing water quality. The proposed method demonstrates that machine learning, particularly ensemble techniques like Random Forests, can serve as effective decision-support tools for water quality monitoring and public health management achieving an overall accuracy 81%..