A Mathematical Epidemiology Approach for Identifying Critical Issues in Social Media

摘要

This work innovated mathematical epidemiology concepts to create a model used to identify critical social issues. A compartmental SEI (acronym for Susceptible Exposed Infected) model was developed to investigate the flow of issues in social media. The basic reproduction number Ro, the number of secondary cases resulting from the introduction of an index case into an otherwise uninfected population, was derived for the model. Any social issue with R_0 greater than one was defined to be a critical issue. The model was evaluated under about 4.7 million tweets from Nigeria streamed over a six-month period and Ro was estimated for each of the top twenty social phenomena in the dataset. Security (R_0 = 30.784), Ebola (R_0 = 51.949), and Transport System (R_0 = 4.166) were found to be critical issues in Nigeria using this methodology.