Water is a natural resource that is very important for the life of living creatures on earth, but water is very easily contaminated with bacteria and dangerous substances. Therefore, it is important to pay attention to the quality of water on earth. To classify water quality as safe or unsafe, there are many methods that can be used. To choose the most suitable method, four methods were used, namely K-Nearest Neighbors (KNN), Naïve Bayes, and Logistic Regression. In this research, the dataset used is Water Quality from the Kaggle website which contains 7999 samples with 20 features and 1 target class. The aim of this research is to compare methods to obtain the highest accuracy values, accuracy results obtained from implementing algorithms in machine learning. The results obtained from the KNN, Naïve Bayes, and Logistic Regression methods were 89.62%, 78.69%, and 89.81% respectively. The highest accuracy result is Logistic Regression, so this method is the best method for classifying water quality data.
                        
                        
                        
                        
                            
                                Copyrights © 2024