AmmarRashed/UnsupervisedStanceDetection

An unsupervised method for target-specific stance detection using embeddings-based clustering and projection techniques. Achieves 90% precision without requiring platform-specific features.

26
/ 100
Experimental

This tool helps researchers and analysts automatically identify different viewpoints or "stances" within large collections of social media posts, like tweets, about specific topics. You provide a file with user IDs and their corresponding posts, and it outputs detected stance groups. It's ideal for social scientists, political analysts, or market researchers studying public opinion and polarization.

No commits in the last 6 months.

Use this if you need to understand divergent opinions on a specific topic from social media data without having to manually label or train a machine learning model.

Not ideal if your data is not text-based or if you require real-time stance detection for streaming data.

social-media-analysis public-opinion political-science market-research sentiment-analysis
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 4 / 25

How are scores calculated?

Stars

24

Forks

1

Language

Jupyter Notebook

License

MIT

Last pushed

Mar 24, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/AmmarRashed/UnsupervisedStanceDetection"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.