Machine Learning Diagnosis of Dengue Fever: A Cost-Effective Approach for Early Detection and Treatment

Abstract

This research aims to explore the potential of machine learning algorithms for diagnosing of dengue fever and assess their cost-effectiveness compared to conventional methods. Four machine learning classifiers (K-Nearest Neighbor, Naïve Bayes, Support Vector Machine, and Random Forest) were utilized. Feature selection and data balancing techniques were employed to enhance algorithm performance. The classifiers achieved high accuracy rates, with Naïve Bayes, Support Vector Machine, and Random Forest achieving 100% accuracy and K-Nearest Neighbor achieving 97.3% accuracy. Additionally, the cost-effectiveness analysis demonstrated that machine learning models for disease classification are the most cost-effective approach due to early detection and diagnosis, resulting in reduced healthcare costs. Therefore, it is recommended to promote the use of machine learning techniques in disease treatment for early detection and improved costeffectiveness.

Keywords

• machine learning classifiers
• dengue fever
• cost-effectiveness