Social media lie detector on its way — will decipher truth from fiction

If the latest statistics are correct, chances are nine out of the 10 people reading this article right now have a social media profile in some capacity. Facebook, Twitter, LinkedIn—heck, if you have a Gmail account then that means you have a Google+ account (even if you didn’t actually set one up). 

Now, whether or not you actually use that profile is a topic of discussion for another article. Chances are, though, that at some point in the past, you’ve scrolled the news feed of your social media channel of choice and exposed yourself to what can best be described as a lot of “chest-thumping” and sensationalized news stories. 

But what’s fact and what’s fiction? A group of researchers set out to create a technology that’ll be able to decipher the difference.

Based out of the University of Sheffield, the system is named after the Greek mythological character Pheme (famed for spreading rumors). Its purpose will be two-fold: to analyze social media feeds in real time to determine whether a posting is true or not, and identify whether or not accounts have been created for the purpose of spreading rumors.

The point of this is to help governments, emergency services, and organizations respond more effectively to events as they’re happening. This makes sense as the idea for a technology like this came from a review on how social media was (mis)used during the London riots in 2011.

The service will analyze four types of online rumor:
• Speculative: like when a new product is set to launch
• Controversial: any political point or counterpoint really
• Misinformation: when lies are spread unknowingly
• Disinformation: when lies are spread maliciously

“There was a suggestion after the 2011 riots that social networks should have been shut down, to prevent the rioters using them to organize,” said 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 rumors, for example, for the emergency services to quash a lie in order to keep a situation calm,” she added.

The system will first categorize sources of information to assess their authority. These categories include:
• News outlets
• Journalists
• Experts
• Eye witnesses
• Members of the public
• Robots that automatically generate posts

The system will also review the account’s history, to determine whether it was created for the sole purpose of spreading rumors.

Next, conversations on social networks will be analyzed to see how they evolve, and from there, sources will be checked to see if information can be confirmed or denied.

“Only text will be analyzed,” said Dr. Bontcheva. “We will not be doing image analysis, so we won't be looking to see if a photo has been altered — it's too technically difficult.”

The results of the system’s analysis will be displayed on a visual dashboard so users can see if the story is true or not.

Everything is set to be up and running in 18 months or so, though at first it’ll only be made available to a small group of journalists and healthcare professionals for testing.

“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,” said Dr. Bontcheva.

Along with the University of Sheffield, there are four other schools involved in developing this system, including Sheffield, Warwick, King's College London, Saarland in Germany, and Modul in Vienna. Four companies are also contributing to its development: Atos, iHub, Ontotext and swissinfo.

Story via: bbc.co.uk