Diabetes is a chronic disease that is considered a significant health problem worldwide. Early detection and prediction of diabetes is a crucial step to enable early intervention and prevent complications. This study aims to apply the Naïve Bayes algorithm in predicting the probability of someone having diabetes. The dataset used in the study was obtained from the National Institute of Diabetes and Digestive and Kidney Diseases. Attributes such as gender, age, body mass index, glucose level, and others were used as independent variables in the Naïve Bayes algorithm to classify them into two groups: having or not having diabetes. From the research results, it has been shown that the Naïve Bayes algorithm can produce a prediction accuracy of 84.6%, 82.3% precision, and 60.8% recall.
Copyrights © 2024