HateBR is the first large-scale, expert-annotated dataset of Brazilian Instagram comments for hate speech detection. It contains 7,000 documents manually annotated by specialists, collected from comments made by politicians. The dataset was last updated on March 9, 2025.
Use Cases
- Train binary classifiers for offensive/non-offensive content based on the described annotation layer.
- Benchmark hate speech detection models on Brazilian Portuguese social media text.
- Analyze linguistic patterns of hate speech in Instagram comments from politicians.
- Develop multi-label classifiers using the three distinct annotation layers mentioned in the description.
Strengths
- 7,000 documents provide a substantial corpus for training and evaluation.
- Expert-annotated labels likely ensure high-quality ground truth.
- Focus on Brazilian Instagram comments offers a specific, real-world data source.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is known, but other metadata like file formats and license are unknown.
- Data may reflect geographic and source bias inherent to comments from Brazilian politicians on Instagram.
Provenance
- Source
- Brazilian Instagram comments made by politicians.
- Collection Method
- Manually annotated by specialists.
- Freshness
- Last updated 2025-03-09 16:04:04; freshness should be verified.
- Geography
- Brazil