Exploring Social Media Misinformation in the COVID-19 Pandemic Using a Convolutional Neural Network

ABSTRACT

Misinformation is rampant in the modern information age and understanding how social media misinformation diffuses can provide vital insight on how to combat it. With social media becoming a major information source, it is increasingly important to address this concern. Social media misinformation has negatively impacted healthcare response in the past and may have played a major role in how to respond to COVID-19. Understanding how misinformation diffuses through online social networks can provide help healthcare and government entities information on how to mitigate the associated negative impact. This paper proposes a data set as criterion for identifying pandemic specific misinformation and develops a Convolution Neural Network model and. A case study is then conducted to illustrate how diffusion can be explored using labelled misinformation. The work shows a decrease of COVID-19 misinformation over time and a pattern that does not depend on regional geographic location. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.