Abstract—Dengue fever outbreaks (DENGUE FEVER) represent a significant public health concern in tropical regions, including South Sumatra Province, where rapid climate changes and complex environmental factors contribute to the unpredictability of outbreaks. This study develops a hybrid predictive model utilizing Support Vector Machine (SVM) and Support Vector Regression (SVR) to identify high-risk areas and forecast the future trends of dengue cases. SVM, known for its classification ability, is employed to precisely classify endemic and non-endemic areas, while SVR, with its capacity to model temporal dynamics, is used to predict the number of future cases based on climate variables, such as temperature, rainfall, and humidity, including time-lagged data to capture delayed environmental effects. The model’s performance was evaluated using real-world data, revealing that integrating SVM and SVR significantly improves both spatial and temporal predictions of DENGUE FEVER outbreaks. SVM’s classification output helps identify areas prone to outbreaks, while SVR provides a detailed forecast of potential case numbers. The model demonstrated high accuracy in mapping endemic zones and predicting case trends, thus addressing both the spatial and temporal aspects of DENGUE FEVER epidemiology. The strength of this approach lies in its ability to process high-dimensional and time-lagged data, providing insights into the delayed effects of environmental factors on disease transmission. The predictive model is valuable for identifying risk areas and assisting health authorities in resource allocation and intervention planning. This study contributes to developing more reliable early warning systems for DENGUE FEVER and lays the groundwork for applying this hybrid machine learning method to other infectious diseases. The results offer significant implications for enhancing preventive measures and public health management in tropical regions.
 
Keywords—dengue fever, Support Vector Machine (SVM), climate variables, time-lagged data, disease prediction


Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).