Optimization of Naïve Bayes Method Using Smoothing and Feature Selection for Dengue Fever Disease

Abstract

Dengue hemorrhagic fever (DHF) is an infectious disease caused by the Dengue virus and has emerged as a significant health issue in many tropical countries, including Indonesia. Early identification of the disease is crucial to prevent further spread and complications. This study aims to refine the Naïve Bayes methodology to improve the accuracy of early detection of medical data related to patients suffering from DHF. The application of Naïve Bayes is expected to enhance predictive accuracy and facilitate healthcare professionals in diagnostic procedures. The data used in this research consists of clinical patient information, including laboratory findings and experienced symptoms. The results show that the optimization of the Naïve Bayes method successfully increased prediction accuracy to 92%, which could serve as an effective diagnostic alternative for early DHF detection. The conclusion of this study is that Naïve Bayes can be relied upon to identify DHF more quickly and accurately, ultimately contributing to the medical decision-making process.