Researchers from the University of Warwick, U.K., have created algorithms to spot fake dating profiles on online dating websites and apps.

According to researchers, the algorithms were developed using artificial intelligence (AI) and are capable of detecting fake dating profiles created for the purpose of bilking victims out of significant sums of money. The algorithms can be used to scan profiles upon their submission to online dating sites and apps, looking specifically for fraudulent information that perpetrators unknowingly include in the demographics section of a dating profile. Likewise, the algorithms will also scan accompanying images and self-descriptions for fraudulent information. So far, the algorithms have performed accurately under testing, producing an extremely low false positive rate — incorrectly identifying just 1% of legitimate dating profiles as fake profiles.

Dating websites and apps are frequented by millions of people worldwide and have also become a hunting-ground of sorts for those looking to scam victims out of money. Often, according to reports, users of the apps and websites can be groomed to give the recipient of their online affection gifts of money or loans that will inevitably go unrepaid. Such exchanges can be both costly and embarrassing for users who tend to not report these encounters.

With funding from the Engineering and Physical Sciences Research Council (EPSRC) and the Economic and Social Research Council (ESRC), the University of Warwick team hopes to make the algorithms available to popular dating websites and apps in the next few years following additional developments to thwart potentially fraudulent dating profiles from being created.

As AI develops, researchers and innovators are finding it useful for a number of unexpected applications. It is being used to create artwork, write fairy tales and it is even being used to help users of dating websites and apps find love.

The study, DAPM: Detecting and Preventing Mass-Marketing Fraud (MMF), appears in UK Research and Innovation.

To contact the author of this article, email mdonlon@globalspec.com