Comparison between Machine Learning algorithms in the prediction of Dengue cases

Dengue is an arbovirus that significantly affects public health, especially in tropical and subtropical regions such as Brazil, where epidemic outbreaks are frequent. Anticipating its spread is essential for preventive actions and agile responses by health services. This study evaluated and compared the performance of machine learning and deep learning algorithms in predicting dengue cases, using climatic and epidemiological variables organized in a temporal manner. Classical models such as Random Forest, Support Vector Regression (SVR) and Multilayer Perceptron were analyzed, as well as deep neural networks such as Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN). The results indicated that the LSTM model presented the best balance between predictive accuracy and execution time, while the SVR stood out as an efficient alternative in scenarios with limited computational resources.