Social Media and Influenza Emergency: Content Analysis of Tweets during the 2015 MERS Outbreak in Korea

Abstract

The risk perception is a key issue for effective health risk communication. This study gathers and analyzes MERS-related tweets in order to monitor and understand the public perception on Twitter during the MERS outbreak of 2015 in South Korea. The main purpose of study is to identify and describe the changes in the number of MERS-related tweets. Using Naïve Bayesian classifiers, this study investigates the public sentiments on Twitter and examines how the sentiments affect the spreading volume, duration, and speed of MERS-related tweets. The results show dramatic changes in the amount of MERS-related tweets during these sequences of time periods. The results of sentiment analysis also showed that the emotional response of the public on Twitter had changed in a variety of ways depending on the types of triggering events. Moreover, an emotional (positive or negative) message in social media seems to receive more attention and feedback than a non-emotional message and induce cognitive and arousal-related effects that trigger sharing behavior in social media communication.