Mosquito-borne diseases represent a significant public health threat in tropical and subtropical regions, with prominent dengue, zika, and chikungunya cases. These three viral infections often exhibit similar clinical symptoms in their early stages, which can lead to misdiagnosis and negative outcomes. While laboratory tests are the most reliable means of accurate diagnosis, such facilities are not always readily accessible in affected areas. Consequently, there have been proposals to leverage Machine Learning techniques to facilitate prompt diagnosis of these diseases. This study applied these techniques to a comprehensive database of dengue, zika, and chikungunya patients, achieving classifiers with an accuracy of 91.1% and performance of 95%, 85%, and 96% for F1-score, Cohen’s kappa, and AUC, respectively. In principle, integrating these techniques could improve accessibility and effectiveness in differential diagnosis.