Application of the Decision Tree C4.5 Method on the Classification of Diet Types of People with Diabetes Mellitus

During the pandemic, people with Diabetes Mellitus are more susceptible to COVID-19. Diabetes Mellitus can cause damage to the heart, blood vessels, nerves, kidneys, and eyes, causing complications. However, the lack of knowledge or understanding of Diabetes Mellitus is an obstacle in the healing process. One way to maintain health and treat Diabetes Mellitus is to adjust the diet, this can be done by dieting regularly. Twelve types of diets are distinguished by calorie levels from the lowest to the highest. The aim of this research is to determine the results of Accuracy, Precision, Recall, dan F1-Score on the type of diet for people with Diabetes Mellitus and to compare the Accuracy, Precision, Recall, dan F1-Score results from the application of the Decision Tree C4.5 method using WEKA software. The variables used to determine the type of diet include weight, height, age, nutritional status, and gender. The research phase begins with a literature study, data collection, data processing, and data analysis to the evaluation stage. The data was obtained from 226 people with Diabetes Mellitus, which were divided into four classifications of types of diet named Diabetes Mellitus IV, V, VI, and VII. The data testing used K-Fold Cross Validation with K-Fold = 10. Decision Tree C4.5 produces Accuracy, Precision, Recall, and F1-Score with accuracy = 86.28%, precision = 86%, recall = 86%, and F1 Score = 86%. From these results, Decision Tree C4.5 is suitable to be used in determining the type of Diabetes Mellitus diet.