A FRAMEWORK OF EPIDEMIC MODELING CONSIDERING PREVENTIVE BEHAVIORS: COMPARTMENTAL MODELING, TEXT ANALYTICS, AND MACHINE LEARNING

Abstract

The goal of this research is to provide a novel framework for epidemic modeling incorporating metrics derived from social media to predict epidemic dynamics and to estimate the impact of preventive behaviors. This study employs empirical data collected from Centers for Diseases Control and Prevention, and Twitter (or X) to demonstrate the practical usability of the proposed framework. Specifically, this research utilizes optimization, simulation, and compartmental differential equations to predict the number of infected and deceased individuals. The research estimates the basic reproduction number (R0) for diseases dynamics. In addition, this study utilizes artificial intelligence and develops a self-training machine learning algorithm to predict the individual compliance level with prevention behaviors. In the analysis, the effect of preventive behaviors on mitigating transmission is evaluated quantitatively. The research contributes to enhance the accuracy of epidemic modeling and to improve decision-making within public healthcare systems, ultimately leading to a reduction in mortality rates and the saving of more lives.