Let's be honest, most of us have told a lie in an email. From pulling sick days to swerving tedious events on your social calendar, spinning tales over email seems like a foolproof method of delivery. Until now, that is, as academics have developed an algorithm that can detect lies in emails.
A research team at the Cass Business School, part of City University London, may have found a way to sniff out deceitful emailers with a computer programme that identifies the use and structure of language in text to detect deception.
Using an archive of emails they studied the multi-faceted aspects of digital language, assessing the ability of word use (micro level), message development (macro level), and intertextual exchange cues (meta level). It might seem complicated, but what they found was relatively straightforward.
If all this rings true, you may want to change your approach: avoid overdoing it and don't try to be best mates with your boss. Or don't lie.
For those who suspect someone might be telling a few porky pies, now you know what to look for.
- Deceitful emailers avoid the use of personal pronouns and superfluous descriptions such as unnecessary adjectives
- Deceitful emailers over-structure their arguments
- Deceitful emailers minimise self-deprecation but include more flattery, and pattern the linguistic style of the recipient across email exchanges because they want to make themselves appear more accommodating and likeable
But the team didn't create this just to catch out people trying to avoid attending an overly enthusiastic friend's book club. The algorithm has been designed to safeguard organisations against fraud and financial loss, as explained in their paper entitled Untangling a Web of Lies: Exploring Automated Detection of Deception in Computer-Mediated Communication.
"This research opens up the possibility of fraud prevention and deception-detection technology across lots of in-person domains. Authorities and companies will now be able to figure out the plausibility of fraud and identify lying individuals. Our approach comes from big data – combining statistics with natural language processing patterns that tip us off to deception," said Dr Tom van Laer, senior lecturer in marketing at the business school.
Using this software, organisations could train managers to identify linguistic cues to spot dishonesty, which could cost companies a lot of money, particularly when it comes to customer complaints or legal claims.
"Our lie-detection software can help companies to assess whether their customers bend the truth in their favour and to decide whether they want to continue doing business with them," said Ko de Ruyter, Professor of Marketing at Cass.
The software, while useful for helping companies to spot a lie, it doesn't offer any guidelines on how to handle deceivers. Instead it may be a served as an alternative form of spam mail that can identify how a lie-ridden email is composed and siphon it away from taking up a company's time.