Russian News Comment Stance Detection Dataset

Aggregating Russian-language comments on news articles, annotated with the comment's stance towards the article. It is designed for stance detection and text classification tasks. The dataset was created by strombergnlp and last updated in October 2022.

Use Cases

• Train a text classification model to predict comment stance using the comment text and article title.
• Analyze linguistic patterns in Russian comment text associated with different stance labels.
• Build a fact-checking or sentiment analysis system for Russian social media discourse based on stance annotations.

Strengths

• Dataset is annotated for stance detection, a specific text classification task.
• Annotations are expert-generated, suggesting a controlled labeling process.
• The dataset is monolingual (Russian), providing a focused resource for that language.

Limitations

• The sample size is categorized as '1 K', indicating it may contain only around 1,000 rows.
• Data is from 2022, which may not reflect current language use or topics.
• Geographic origin is listed as 'Region:us', which may indicate a non-Russian source for Russian comments, potentially introducing bias.

Provenance

Source

strombergnlp

Collection Method

Contains comments on news articles with expert-generated stance annotations.

Freshness

Last updated 2022-10-25.

Geography

Region indicated as 'us', but data language is Russian.