COVID-19 Social Media Posts: 9 Million Tweet IDs from January to April 2020

A collection of 9 million tweet IDs related to the COVID-19 pandemic, published between January 27 and April 20, 2020. The dataset was created by Negin Karisani of Purdue University West Lafayette and includes a pre-trained BERT model for text analysis.

Use Cases

• Train or fine-tune language models for pandemic-related text based on the included BERT model.
• Conduct longitudinal sentiment analysis of public reaction to COVID-19 based on the tweet collection timeframe.
• Study information diffusion and topic evolution on social media during a global health crisis.

Strengths

• Contains 9 million tweet IDs, providing a large-scale sample of pandemic-related social media data.
• Covers a specific and critical three-month period (Jan 27 - Apr 20, 2020) during the initial global outbreak.
• Includes a pre-trained BERT model, offering a starting point for domain-specific NLP tasks.

Limitations

• Row count for the primary tweet data is unknown; the 9 million figure refers to IDs, not the full text content.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

paperswithcode

Collection Method

Tweet IDs were collected and mentioned in an associated academic paper.

Time Range

January 27, 2020 to April 20, III

Freshness

null

Geography

null