Building an Open-source Database of Misinformation Sources on COVID-19

Posted on:  2020-04-02

As the novel coronavirus continues to spread throughout the world, so does false or misleading information about the pandemic online. The World Health Organization (WHO) Director-General Tedros Adhanom Ghebreyesus went so far as to say that “we’re not just fighting an epidemic; we’re fighting an infodemic.”

In this context, fact-checkers from around the world have been working around the clock to check the veracity of rumors regarding COVID-19. To help organize this collective effort, the International Fact-checking Network (IFCN) launched the CoronaVirusFacts Alliance in January 2020, creating a database that gathers all of the fact-checks on the current epidemic in one central location. The fact checks come from more than 100 fact-checking organizations in 45 countries and are written in at least 15 different languages.

The CoronaVirusFacts (CVF) database contains the urls of some articles containing false or misleading claims that have been fact-checked. However, we know that false claims are very often repeated on dozens more websites and hundreds to thousands of social media posts, as was the case last year with a fake “cure for cancer” claim that was first shared by the Jerusalem Post.

Figure – Network of blogs that have published at least one article on COVID-19 rated ‘False’ by Science Feedback (red dots) and Facebook pages that have shared at least two of these urls (blue dots; size proportional to number of followers). This network is based on a dozen fact-checks from Science Feedback; it gives a hint of what we can build with several thousand fact-checks from around the world!

The goal of the project, led by Science Feedback, is to now expand the CVF database to include all of the urls in which an article repeated and endorsed COVID-19 misinformation. Through a collective effort among citizen scientists, volunteer programmers, and fact-checkers, this project will produce a database of websites and social media accounts that are most active in spreading COVID-19 misinformation. The database will prove useful to scientists investigating misinformation networks, to social media companies, and to search engines that attempt to elevate reliable content during epidemics in order to prevent their recommendation algorithms from promoting misinformation.

This project is supported in part by the Google News Initiative. If you’d like to help us make the Internet a more reliable source of credible information during this pandemic and beyond, please join our effort! Here are a few ways:

  • Email us if you are a web developer and want to help write code for our platform (Postgres DB, Python/Flask backend, React front-end).
  • Donate to support this project.

Related Articles