Communicable disease surveillance through predictive analysis: A comparative analysis of prediction models

Abstract

Effective prediction and surveillance of communicable diseases are vital for public health management. This study leveraged machine learning algorithms to predict disease occurrences in the Province of Marinduque, focusing on Hand Foot Mouth Disease, Dengue, Typhoid, Influenza, Chikungunya, Rabies, Measles, Meningitis, Hepatitis, and Acute Bloody Diarrhea using data from 2015 to 2019. The monthly morbidity rate served as the criterion variable. Machine learning models, including Random Forest, Logistic Regression, SVM, and k-Nearest Neighbors, were employed. Material and methods encompassed data collection, preprocessing, feature selection, and model evaluation. Results revealed Random Forest as the most accurate algorithm, with implications for proactive disease management and resource allocation. This research enhances disease prediction methodologies and contributes to public health surveillance.