Computing Veracity – the Fourth Challenge of Big Data

Pheme work on computing veracity wins Best Paper prize

Work on measuring how true a document is from the words it uses has won the best paper prize at a major workshop, W-NUT. This is the workshop on noisy, user-generated text, hosted in 2016 at the conference for Computational Linguistics (COLING) in Osaka, Japan.

In this research, AI methods for determining how certain one could be about the claims made in documents were built from scratch, and tested on stories around three of the PHEME use-cases: the Ottawa Parliament Hill shooting, the Sydney hostage crisis, and the Germanwings suicide crash. These methods then fed into a system for finding which tweets would resolve a rumour, and which way it would go, with 74% and 76% accuracy respectively.

The paper, “Veracity Computing from Lexical Cues and Perceived Certainty Trends”, by Uwe D. Reichel and Piroska Lendvai, can be read here:

Be Sociable, Share!

Comments are currently closed.