Viral US Election Tweets Labeled for Fake News, November 2016 to March 2017

A collection of tweets related to the 2016 US election that went viral between election day and March 2017. The dataset was created by Julio Amador of Imperial College London, querying Twitter's streaming API using specific hashtags and user handles. Tweets are labeled for containing fake news, defined as serious fabrication, large-scale hoaxes, jokes taken at face value, slanted reporting, or contentious stories.

Use Cases

• Detect fake news in political discourse based on the provided fake news labels.
• Analyze the spread of viral content during a major election period based on the retweet threshold.
• Study the characteristics of tweets flagged as hoaxes or slanted reporting based on the described labeling criteria.
• Train classifiers to identify fabricated or contentious stories in social media text.

Strengths

• Tweets are labeled for fake news by two sets of people, suggesting a multi-rater annotation process.
• Focuses on viral content, defined as tweets achieving a 1000-retweet threshold during the collection period.
• Collection uses specific, election-relevant hashtags and user handles for targeted data gathering.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Imperial College London

Collection Method

Collected via Twitter's streaming API using hashtags #MyVote2016, #ElectionDay, #electionnight, and user handles @realDonaldTrump and @HillaryClinton.

Time Range

November 8, 2016 to March 2017

Geography

United States