Researchers Create The First Twitter Lie Detector

twitter lie detector

Liar, liar, pants on fire. If you’re thinking about spreading misinformation on Twitter, think again: researchers are building a Twitter lie detector to catch unsavory types in their 140 character lies.

In an international effort, researchers have banded together to create Pheme, a system to determine whether someone is telling the truth on Twitter.

As The Telegraph reports, the European Union is supporting the £3.5 million research project.

The Pheme system would be able to test information on Twitter quickly, and trace a story back to its origin. Researchers say this type of technology would have been useful during an event like the London Riots of 2011, when Twitter users spread false information about animals released from the London Zoo and landmarks set on fire.

The system will classify tweets that it deems to be false into four categories: speculation, controversy, misinformation (the spreading of lies unwittingly) and disinformation (the spreading of lies with malicious intent).

In order to verify information, Pheme will search for sources that can back up claims, and map out how a certain piece of information spread across the social web. The software will also dig into the background of the users sending the tweets to determine whether they are bots created for the sole purpose of spreading incorrect information.

Pheme is expected to be up and running in about 18 months, with early versions being made available even sooner.

(Liar image via Shutterstock)