Dengue Fever Prediction Using Machine Learning Approaches

Abstract

Dengue fever, also a viral infection spread by mosquitoes, is still a major global health concern, impacting millions of people each year. By applying a carefully controlled dataset of 521 entries and 23 variables, this study analyzes the predictive efficacy of various machine learning methods for Dengue Fever. Among the methods tested, SVM outperforms the others, obtaining an excellent accuracy of 98.88%. This remarkable accuracy highlights the algorithm's ability to capture complex patterns within the multidimensional dataset, establishing it as a strong choice for Dengue Fever detection. MLPclassifier comes in second with an impressive accuracy of 97.78%, indicating its ability to handle the dataset's constant characteristics. The success rate of Logistic Regression and GaussianNBis 96.95% and 93.64%, respectively, illustrating how they adjust to the dataset's complexities. BernoulliNB, on the other hand, lags behind with a lower accuracy of 67.85%, indicating limits in dealing with the dataset's peculiarities, particularly given its affinity for binary features. SVM exceptional accuracy highlights its promise as a significant tool for effective Dengue Fever detection. The study provides essential knowledge for health professionals and academics, guiding the selection of the most successful modeling algorithms in the context of infectious diseases.