Diabetes Mellitus is a disease characterized by an increase in glucose as well as an abnormal rise in blood sugar concentration due to insulin deficiency. The International Diabetes Federation (IDF) reports that in 2021, approximately 540 million people worldwide were affected by diabetes, and this number is expected to increase further if the general public's lack of awareness about symptoms that can trigger the diabetes disease continues. This research aims to implement the C4.5 algorithm in predicting diabetes mellitus based on acquired data. The amount of data used is 300 records, where 90% of the data serves as training data and the remaining 10% is test data. The data consists of 6 attributes: age, gender, hypertension, glucose, heart disease, and BMI (Body Mass Index). Based on Gain calculations, the Glucose attribute becomes the root of the decision tree. The tested data in this study achieved an accuracy rate of 77%, precision of 82%, and recall of 64%.
Copyrights © 2023