Crisis Management Tools: A Lie Detector for Tweets

Will a program be able to sort fact from rumor on social media?

Social media rumors have created crises not only for those who find themselves the focus of said rumors, but also those who propagate the untruths. Now, European researchers say they have created a system which would help Twitter users discern exactly how reliable any given 140-character message is.

The UK Telegraph’s Keith Perry shared details on the new “lie detector for tweets”:

Pheme will classify online rumours into four types: speculation, such as whether interest rates might rise; controversy — such as the heated debate over the MMR vaccine; misinformation, where something untrue is spread unwittingly; and disinformation, where false statements are published with sinister motives.

The system works by assessing the quality of information and sources, giving more weight to established news outlets and experts, and looking out for “bots” (computer programs designed to send spam). It will also look for a history and background of users, so that it can spot where Twitter accounts have been created purely to spread false information.

Pheme will then search for sources that can back up or dismiss the information, as well plot how the conversations on social networks evolve, using all of this information to assess whether it is true or false. The results will be displayed to the user on screen, telling people if an untruth is taking hold among the public.

This is a lovely idea in concept, but we have serious reservations about whether it would be possible to make such a system reliable enough to be employed as a crisis management tool. We anticipate not only the expected technical issues, but also clever users figuring out how to game the system shortly after the “lie detector” goes live, putting anyone depending on its output to protect their own reputation at risk.

The BCM Blogging Team
https://www.bernsteincrisismanagement.com

Leave a Reply