Exploring an Effective Machine Learning Method for Dengue Fever Prediction

This study aims to build models based on the spread of dengue fever to predict its epidemic trends in different regions. Dengue fever is a mosquito-borne disease. Climate change, such as temperature and precipitation, is closely related to its spread, which is a major concern for public health in recent years. Taking the cities of San Juan and Iquitos as examples, this study uses machine learning to predict the trend. The model development tried methods such as random forest regression, KNN, XGBoost, LSTM, and support vector regression. The XGBoost performed best for San Juan while SVR excelled for Iquitos.