several27/FakeNewsCorpus
A labeled corpus of millions of news articles for fake news detection training.
★413 stars Data Tooling
Not currently ranked — collecting fresh signals.
star history
This repository hosts a large-scale news article dataset scraped from over 1000 domains, curated for fake news recognition tasks. Articles are labeled by source type (fake, bias, reliable, etc.) and include extracted metadata such as title, authors, content, and keywords. The corpus is formatted as CSV and intended for training NLP and deep learning classification models.
Frequently asked
- What is several27/FakeNewsCorpus?
- A labeled corpus of millions of news articles for fake news detection training.
- Is FakeNewsCorpus open source?
- Yes — several27/FakeNewsCorpus is open source, released under the Apache-2.0 license.
- How popular is FakeNewsCorpus?
- several27/FakeNewsCorpus has 413 stars on GitHub.
- Where can I find FakeNewsCorpus?
- several27/FakeNewsCorpus is on GitHub at https://github.com/several27/FakeNewsCorpus.