9,008 Danish sentences labeled with fine-grained sentiment scores ranging from -2 to 2. The dataset focuses on political commentary and provides both raw fine-grained polarity and cross-validated simple polarity labels.
Use Cases
- Train a regression model to predict sentiment intensity using the -2 to 2 polarity scale
- Develop a binary sentiment classifier for Danish text using the cross-validated simple polarity labels
- Perform linguistic analysis on Danish political speech by correlating sentence content with polarity scores
Strengths
- 9,008 sentences extracted from Danish political comments
- Fine-grained polarity labels ranging from -2 (negative) to 2 (positive)
- Cross-validated simple polarity labels included to address uncertainties in fine-grained scoring