Skip to main content
. 2021 Apr 14;7(4):e26527. doi: 10.2196/26527

Figure 2.

Figure 2

Tweet-filtering flow. The initial tweet corpus was obtained from Chen et al [41], who used keywords and known accounts to provide a sample of COVID-19–related Twitter data (Filter 1). We then used regular expressions to create four conspiracy theory data sets (Filter 2) and machine learning classifiers to identify misinformation tweets within each data set (Filter 3). 5G: conspiracy theories related to 5G technology; CDC: US Centers for Disease Control and Prevention; Gates: conspiracy theories related to Bill Gates or the Bill & Melinda Gates Foundation, Lab: conspiracy theories related to the virus being laboratory-released or human-made; Vax: conspiracy theories related to vaccines; WHO: World Health Organization.