Called ‘Pheme’, it will scan tweets from news outlets, individuals, and automated ‘bots’ to see if their posts are as truthful as they claim.
Such a system could have been valuable for situations like the 2011 London riots, where rioters used networks like twitter to organise themselves and spread untrue information, said scientist Kalina Boncheva of the University of Sheffield in Britain.
The lie detector is named after a Greek goddess famed for spreading rumours.
Social networks also provide useful information – the problem is that it all happens so fast and we can’t quickly sort truth from lies.
This system aims to help with that by tracking and verifying information in real time.
Pheme would use a range of different indicators to tell the difference between truthful tweets, malicious rumours and harmless boasts.
It would trawl through the history and background of users to pinpoint accounts that were created purely to spread lies and rumours, said a report on Daily Mail.
It would classify online rumours into four different types – speculation, controversy, misinformation and disinformation.