Implementation of GridSearchCV to Find the Best Hyperparameter Combination for Classification Model Algorithm in Predicting Water Potability

Drinking water quality is an important factor in public health, so an accurate approach is needed to determine water potability. This research aims to create a water potability prediction model using machine learning methods, with a focus on model accuracy and testing. The dataset used includes various chemical parameters, as well as one radiological and acceptability parameter. In this study, various machine learning algorithms, such as Support Vector Machine (SVM), Random Forest (RF), and Logistic Regression, were applied using GridSearchCV and their performance compared. Models were evaluated using accuracy, precision, recall, F1-score, and confusion matrix metrics, with cross-validation to ensure generalizability. The results showed that the Support Vector Machine algorithm provided the best performance with an accuracy of 70.43%, followed by Random Forest and Logistic Regression with accuracies of 70.12% and 62.20%, respectively. The Support Vector Machine-based model is able to provide reliable predictions and can be used as a tool to support decision-making in water quality management.