It’s no secret that Facebook has a major problem with fake accounts. In fact, over a three-month period, the social media giant can remove as many of these profiles as there are real ones on its platform. That’s why Facebook introduced a new machine learning tool called Deep Entity Classification (DEC) to automatically detect these profiles. According to Facebook, the system has already reduced the volume of fake accounts on the platform by 27 percent. Additionally, it blocks millions of additional attempts per day to create these spam accounts.
The tool works by going a level deeper than previous spam detectors, identifying “deep features” of an account’s behavior to make it difficult for cybercriminals to game the system. Of course, these scammers are always seeking an edge to stay a step ahead of Facebook.
“The problem is adversarial robustness,” said Facebook data science manager Bochra Gharbaoui. “Adversaries can control direct features quite easily, for example by managing the number of friend requests they send out. That’s why you need to look at deep features — the graph and the model around the account.”
With a platform as large as Facebook, a 27 percent reduction in fake accounts is hugely significant. Unfortunately, it’s likely a problem that can never be fully solved.
The Choice of Tech Experts Worldwide. Try 90 days free of Bitdefender 2020 and experience the highest level of digital safety.
Private Internet Access is an award-winning, cost-effective VPN solution. The use of an anonymous and trusted VPN is essential to your online privacy, security and identity protection.