Ensemble Learning dengan Soft Voting Classifier untuk Klasifikasi Pasien Tifus di Puskesmas Balung

Typhoid fever remains a significant public health problem in Indonesia, particularly in areas with limited medical facilities. This study aims to develop an automatic classification model for typhoid diagnosis using an ensemble learning approach based on the Soft Voting Classifier. The model combines three base algorithms, Logistic Regression, Random Forest, and Gradient Boosting, to enhance predictive accuracy. The dataset was obtained from Balung Primary Health Center, Jember Regency, consisting of 510 patient records with typhoid symptoms. Experimental results show that the ensemble model achieved an accuracy of over 92%, outperforming individual models. Furthermore, adequate precision and recall indicate the model’s potential to support rapid and accurate medical diagnosis. These findings demonstrate that the Soft Voting Classifier can serve as an effective tool to assist healthcare workers, especially in resource-limited settings, in improving the quality of typhoid fever diagnosis.