This dataset aggregates fact-checked claims from global news media organizations as indexed by Data Commons. It provides structured metadata including the original claim text, the entity making the claim, and the truthfulness rating assigned by professional fact-checkers.
Use Cases
- Train natural language processing models for automated claim detection using the claimReviewed text
- Analyze the distribution of misinformation across different publishers using the author and publisher fields
- Develop stance detection algorithms by comparing claimReviewed strings against their corresponding reviewRating
- Map the spread of specific narratives by filtering claims by datePublished and locationCreated attributes
Strengths
- Contains structured ClaimReview metadata including claimReviewed and reviewRating fields
- Aggregates data from multiple global news media organizations and independent fact-checking bodies
- Includes temporal data via datePublished fields for tracking misinformation trends over time
- Provides source attribution through author and itemReviewed fields