Weather Data Driven Dengue Risk Zone Prediction using Machine Learning Algorithms

Dengue fever remains a serious threat to public health in Bangladesh and in many other tropical and subtropical locations. The prediction of dengue risk zones depends on early and accurate prediction of dengue risk zones. This study follows machine learning techniques to analyze weather data including temperature, humidity, precipitation, and wind speed to predict dengue risk levels. It follows a methodology of data preprocessing, model training using accuracy, precision, recall and F1 score. Several models including Support Vector Machine, Random Forest as well as Gradient Boosting have been applied and compared. The results have shown that the Gradient Boosting has outperformed all the models with an accuracy of 99.54%. The study shows how machine learning can improve dengue risk zone prediction and support evidence-based public health decisions.