‘Given a claim in one post or story and other posts and stories on the same subject that have been collected for comparison, the system can correctly determine if they support it or not nine out of 10 times.’

Researchers at Waterloo University are using a technique known as ‘stance detection’ to ferret out false news reports. Stance detection compares the contents of similar stories to determine truthfulness.

High coherence between stories indicates the story is real; little or low coherence points to fakery.

They claim 90% accuracy in detection.


New tool uses AI to flag fake news for media fact-checkers
SCIENCE DAILY | December 16, 2019| by University of Waterloo