Sistem Prediksi Penyakit Diabetes Berbasis Decision Tree

Data diabetics increased from year to year. Diabetes diagnosis rate contributes signi ficantly to the level of comorbidities and complications of diabetes. Based on historical data diabetics can made recommendations that help predict diabetes health professionals. Classification is one of data mining techniques that can be used to make predictions. Classification can be done with a decision tree, one of them with the C4.5 algorithm. This study to create a data classification of diabetes and apply them in the development of diabetes prediction system. Diabetes data classification results are evaluated by confusion matrix and ROC (Receiver Operating Characteristic)curves to determine the accuracy of the classification results. Evaluation results have shown that including Excellent Classification. Rule classification results are implement ed for prediction system of diabetes. The system is built using Microsoft Visual Basic 6.0 and MySQL database.