Water Quality Prediction and Suggesting Filter using Machine Learning

Survival of all living beings and efficient working of industries and agriculture are largely dependent on the availability of clean water. However, contamination is a major issue that requires dependable methods for assessing the quality of water before its use in drinking, chemical applications, or animal hydration. This research utilizes machine learning to predict water quality based on parameters such as pH, turbidity, hardness, conductivity, and dissolved particles. Based on a broad dataset, models such as time-series analysis and Random Forest classifier are applied to categorize water as safe or unsafe for domestic use. A hardware system equipped with sensors captures real-time data, which enables dynamic assessment of water quality. The proposed solution shows high accuracy in its evaluation using metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Regression Analysis, offering a practical approach to ensure water safety in diverse applications.