Prediction of Water Potability Through Voting Classifier

The conventional individual in good health drinks 2.5 to 3.5 Liters of water per day. A person can stay alive for a few weeks by drinking just pure water while starving. The key objective of this research is to determine the best approach for the statistically imputed dataset and algorithm that yields the most accurate water level prediction. In order to accomplish the best accuracy when compared to the separate algorithms, we integrated two machine learning algorithms using voting classifier. Our dataset, which included nine features and some missing values, was pre-processed using the mean of each characteristic, producing a desirable dataset. Our machine learning algorithms research on water potability yielded an impressive accuracy rate of 91%, as we found in the end. The results reported here demonstrate how machine learning may enhance the evaluation of water quality, providing promising advancements for ensuring consistent and secure water sources.