<![CDATA[Dengue Hemorrhagic Fever (DHF) remains a significant public health issue in Purwokerto, with the increasing number of cases influenced by environmental factors such as temperature, humidity, rainfall, and population density. Accurate and adaptive predictive methods are essential to anticipate the spread of DHF, one of which involves the application of machine learning algorithms. This study aims to compare the performance of three algorithms, namely Naïve Bayes, Binary Logistic Regression, and Support Vector Machine (SVM), in predicting DHF risk in Purwokerto. Secondary data were obtained from the Health Office, Meteorology Agency (BMKG), and Statistics Bureau (BPS), covering DHF case records and environmental factors. The analysis was conducted using a quantitative predictive approach, employing 5-Fold Cross Validation and evaluation metrics including accuracy, precision, recall, and F1-score. The results indicate that the SVM model demonstrated the highest performance with an accuracy of 82% and a high recall rate for the positive class, making it effective for DHF risk mapping. The Naïve Bayes model showed adequate sensitivity but lower precision, while the Binary Logistic Regression model produced the lowest overall performance. This study recommends SVM as the most effective algorithm to support early warning systems and risk mitigation for DHF based on environmental data in Purwokerto.]]>
Copyrights © 2025
