Personal Democracy Plus Our premium content network. LEARN MORE You are not logged in. LOG IN NOW >

Building an Automatic "Lie Detector" for Twitter

BY Jessica McKenzie | Thursday, February 20 2014

American inventor Leonarde Keeler testing his lie-detector (Wikipedia/public domain)

An international group of researchers led by the University of Sheffield is building a social media “lie detector” named Pheme, after the mythological rumormonger, that can determine in real time whether a information spread on social media is true or false.

The idea is that identifying misinformation would allow journalists, government agencies, emergency response, health providers and private companies to respond to emergencies and other events more effectively.

“There was a suggestion after the 2011 riots that social networks should have been shut down, to prevent the rioters using them to organise [sic],” says Dr. Kalina Bontcheva, lead researcher on the project. “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 [sic], for example, for the emergency services to quash a lie in order to keep a situation calm. Our system aims to help with that, by tracking and verifying information in real time.”

It is unclear from the press release how online rumors might be “quashed”—whether they would be deleted from the platform or merely countered with accurate information. The possibility of deletion conjures up specters of censorship in countries like China and Venezuela, where dissidents are silenced for voicing “rumors.”

Much of the technology to distinguish between truth and falsehood is already there, but the challenge is doing it in real time.

Dr. Bontcheva explains:

We can already handle many of the challenges involved, such as the sheer volume of information in social networks, the speed at which it appears and the variety of forms, from tweets, to videos, pictures and blog posts. But it’s currently not possible to automatically analyse, in real time, whether a piece of information is true or false and this is what we’ve now set out to achieve.

Online rumors will be classified into four groups: speculations, controversy, misinformation (the inadvertent spread of a lie) and disinformation (the intentional spread of a lie).

It will also account for the authority of the source, distinguishing between media outlets, individual journalists, experts or other authority figures, potential eye witnesses and members of the general public.

The Swiss Broadcasting Corporation, swissinfo.ch, one of the project collaborators, will be testing Pheme for use in digital journalism.

Other collaborators include Atos, the Spanish ICT company; iHub, the technology research center in Kenya; and Ontotext, a Bulgarian company that builds core semantic technologies.

Personal Democracy Media is grateful to the Omidyar Network and the UN Foundation for their generous support of techPresident's WeGov section.

For a round-up of our weekly stories, subscribe to the WeGov mailing list.

News Briefs

RSS Feed thursday >

NYC Open Data Advocates Focus on Quality And Value Over Quantity

The New York City Department of Information Technology and Telecommunications plans to publish more than double the amount of datasets this year than it published to the portal last year, new Commissioner Anne Roest wrote last week in an annual report mandated by the city's open data law, with 135 datasets scheduled to be released this year, and almost 100 more to come in 2015. But as preparations are underway for City Council open data oversight hearings in the fall, what matters more to advocates than the absolute number of the datasets is their quality. GO

Civic Tech and Engagement: Announcing a New Series on What Makes it "Thick"

Announcing a new series of feature articles that we will be publishing over the next several months, thanks to the support of the Rita Allen Foundation. Our focus is on digitally-enabled civic engagement, and in particular, how and under what conditions "thick" digital civic engagement occurs. What we're after is answers to this question: When does a tech tool or platform enable actual people to make ongoing and significant contributions to each other, to a place or cause, at a scale that produces demonstrable change? GO

More