Computers catch catfishes

Computers are trained to spot social media users who pretend they are somebody else, with 90% accurate algorithms identifying users who lie about their gender.

Pornhub employees checked data from 5000 public profiles manually and showed algorithms of men and women all different ages interacting differently through post comments and writing styles, in a bid to search those lying about gender and age on other websites.

‘Women are more likely to deceive than men’

According to the study, women are more likely to deceive than men, with 40% of users lying about their age and 25% lying about their gender.

“Adult websites are populated by users who claim to be other than who they are, so these are a perfect testing ground for techniques that identify catfishes,” said Dr Walid Magdy, of the University of Edinburgh’s School of Informatics.

“What was interesting was that it seems that for many the reason for lying was to get more friends and subscribers,” he added.

According to consumer group Which, the majority of dating website users have encountered at least one fake profile, with people deceived by dating scams reaching the highest in 2016.

Edited from source by Leah Alger

Source: BBC

More
articles

Menu