Klasifikasi Indikator Kesehatan Diabetes Menggunakan Algoritma Random Forest

Abstract

Diabetes continues to rise as a global health concern, highlighting the need for analytical methods that can assist in earlier and more accurate detection. This study aims to classify diabetes conditions using the Random Forest algorithm implemented through the Orange Data Mining platform. The dataset used contains various health-related attributes such as glucose levels, blood pressure, body mass index, age, and other clinical indicators associated with diabetes risk. Random Forest was selected due to its ability to produce stable models, handle large and complex datasets, and minimize overfitting by combining multiple decision trees. The research process includes data preprocessing, splitting the dataset into training and testing portions, building the Random Forest model, and evaluating its performance using metrics such as accuracy, precision, recall, F1-score, and confusion matrix. The results indicate that Random Forest delivers strong and consistent performance in classifying diabetes conditions based on the given health indicators. These findings suggest that employing data mining techniques especially Random Forest within Orange—can serve as a practical and reliable approach to support medical analysis and assist healthcare practitioners in achieving earlier and more accurate diabetes detection.