FakeHealth: Health News with Social Engagements and User Networks

FakeHealth addresses challenges in fake health news detection by including news content, reviews, social engagements, and user networks. The repository stores Twitter IDs for social engagements and user networks, supplemented with an API to retrieve the full data. It was collected by Dai Enyan of Pennsylvania State University.

Use Cases

• Detect fake health news based on news content and review text.
• Analyze patterns of misinformation spread based on social engagement data.
• Model user influence and network structure in health news dissemination.
• Train multimodal classifiers using text and graph features mentioned in the description.

Strengths

• Includes multiple data modalities: news content, reviews, social engagements, and user networks.
• Provides an API to retrieve full social engagement data from stored Twitter IDs.
• Explicitly designed to address challenges in fake health news detection.

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

Pennsylvania State University

Collection Method

Collected from Twitter, with news contents and reviews.