Social Media Polygraph Finds Truth on Twitter

The world of social media is one fraught with speculation, discussion, debate and well, large helpings of utter horsesh*t.

As an ardent advocate of all mediums of social media I am most certainly guilty of contributing to the genre of misinformation – so am also among the anxious many about the advent of an online lie-detector, specifically being built to weed deceit out of the digital ether.

Currently under development by some time-rich boffins at the University of Sheffield, the insight is that social media holds an as yet untapped potential for unadulterated formal news sharing currently blemished by rumour and speculation. What’s more, the ambition is that a social media polygraph will one day evaluate, dissect and translate the nature of online posts in real-time.

Both an invaluable tool and double edged sword for any brand, newspaper, magazine, blog or governmental communications departments, the state of technology is very much in the research phase, currently one that involves utilising data from the media blasts communicated during the London riots in 2011. By looking at the various posts on Twitter, online forum commentary and public opinions posted on Facebook, the clever algorithms categorise internet rumours into four distinct groups: ‘Speculation’, ‘Controversy’, ‘Misinformation’ and ‘Disinformation’. From here, the researchers say they can more easily disseminate and identify which of them correlate authority in relation to their source, ranging from the disparate informants of journalists, blogs and even automated online bots.

liedetector-1

Adding yet another layer of scrutiny, the software will look at posts from a number of perspectives, even examining an account’s history to understand its motivations as well as corroborating it with contextual support from other online sources.

Dr Kalina Bontcheva, lead researcher on the project at the University of Sheffield said:

“There was a suggestion after the 2011 riots that social networks should have been shut down, to prevent the rioters using them to organise. But 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 makes it difficult to respond to rumours, for example, for the emergency services to quash a lie in order to keep a situation calm”

“We’ve got so see what works and what doesn’t, and to see if we’ve got the balance right between automation and human analysis,”

Named after Greek mythologies ‘Pheme’ – the fabled rumour spreader – is being backed by four other universities in the form of Warwick, King’s College London, Saarland in Germany and Modul in Vienna, as well as companies such as Atos, iHub, Ontotext and swissinfo.

However, what’s most amazing of all is that the initial set of results from this frankly zany pilot scheme is that  results will be expected in as little as a year and  half, with academics saying they’ll create a ‘visual dashboard’ to see if a specific rumour is begging to gain traction.

Advertisements