COVID-19 and Social Media

Last Update: Dec 11, 2020

This page contains resources associated with multiple studies (completed and ongoing) focusing on COVID-19 (scroll down).

Dashboards

Our COVID-19 Twitter dashboards are available now (while under construction; updated 12/21/20):

GLOBAL DASHBOARD

US DASHBOARD

Research topics/studies:

  • Syndromic surveillance: we are utilizing real-time social media data, natural language processing and machine learning methods to identify and track symptom distributions over time.
  • Localized outbreak detection: we are combining retrospective data from social media with data from GDPH to train AI algorithms that can detect patterns in social media chatter indicative of localized outbreaks.
  • Toxicosurveillance: we are building methods that can detect unapproved treatments that are promoted for treating COVID to identify potential toxic exposures.
  • MOUD treatment access during COVID: in line with our past work, we are studying how COVID is affecting SUD and OUD treatment programs.
  • COVID and mental health: we are studying the impact of COVID on mental health using social media mining methods.
  • Impact of COVID on national services such as Medicaid.

    • Publications

    [1] Sarker A, Lakamana S, Hogg-Bremer W, Xie A, Al-Garadi MA, Yang YC. Self-reported COVID-19 symptoms on Twitter: an analysis and a research resource. J Am Med Inform Assoc. 2020 Aug 1;27(8):1310-1315. doi: 10.1093/jamia/ocaa116. PMID: 32620975; PMCID: PMC7337747. [ Paper ]

    Resources:

    Twitter Symptom Lexicon

    Twitter COVID-19 Symptom Lexicon

    Automatically Expanded Symptom Lexicon

    Expanded Lexicon (automatically; via a data-driven method) Symptom Lexicon [Publication Forthcoming].

    COVID19 Word2Vec Models

    Four covid-19-related models (Available here):

  • Small model on Twitter
  • Large model on Twitter
  • Social media model (Reddit + Twitter)
  • Combined dataset (social media & literature)

    • Funding, Collaborations and Acknowledgments

    The studies are currently funded partially or fully by Emory University, the Centers for Disease Control and Prevention (CDC), and the National Institute on Drug Abuse (NIDA).

    The above studies are in collaboration with researchers from Emory University (Medicine, Public Health & Nursing), Georgia Department of Public Health, the Centers for Disease Control and Prevention (CDC), and various universities across the United States.