<![CDATA[This study aims to develop a risk prediction model for Diabetes Mellitus by applying the Decision Tree C4.5 algorithm using the CRISP-DM (Cross-Industry Standard Process for Data Mining) approach. The dataset used includes data on patients diagnosed and undiagnosed with diabetes, with several important medical attributes such as glucose levels, blood pressure, body mass index, age, and family history [1]. Of these attributes, glucose levels have been shown to be the most dominant factor in distinguishing patients at risk from those without [2]. The data was divided into two parts: 80% for model training and 20% for testing. The evaluation results showed that the model produced an accuracy of 79.3%, a precision of 81.0%, and a recall of 76.5% [3]. This indicates that the model is quite effective in identifying patients at risk of Diabetes Mellitus. However, further optimization, such as attribute enrichment and advanced data processing, is still needed to improve the reliability of the predictive model [4]. The resulting model is expected to be a tool in supporting medical decision making, especially in early diagnosis and preventive measures against diabetes [5]. This approach can also encourage increased public awareness of the importance of regular health monitoring.]]>
Copyrights © 2025
