COVID-19 Twitter Data: Annotated Tweet IDs for Public Health Research

Xiaolei Huang at the University of Colorado Boulder provides a dataset of Twitter IDs related to COVID-19. The collection includes the date, inferred geolocation, and keywords for each tweet, enabling the download of original content from Twitter. This resource supports research into public discourse and information spread during the pandemic.

Use Cases

• Tracking public sentiment and discourse evolution based on tweet dates and keywords.
• Analyzing geographic patterns of COVID-19 discussion using inferred geolocation data.
• Studying information diffusion and keyword co-occurrence within pandemic-related tweets.

Strengths

• Provides direct access to original Twitter data via tweet IDs, adhering to platform terms.
• Includes automated annotations for date, keywords, and geolocation, adding analytical value.
• Has an Open Access license, facilitating reuse and distribution.

Limitations

• Key metadata such as row count, file size, and last update date are not provided across all sources.
• The dataset's completeness and temporal coverage are unclear from the available descriptions.
• Relies on inferred geolocation, which may introduce inaccuracies compared to user-provided location data.

Provenance

Source

Xiaolei Huang, University of Colorado Boulder

Collection Method

Collection of Twitter IDs related to COVID-19, with automated annotation for date, keywords, and geolocation.

Geography

Global (inferred from geolocation annotations)