Researchers at the Carnegie Mellon University have developed an algorithm that can detect online fraudulent personalities and behaviors.
The algorithm is called FRAUDAR and it can detect if someone has faked an Amazon or Yelp review or if a celebrity or a politician with a suspiciously large number of Twitter followers might have paid for that popularity.
The researchers used a massive Twitter database extracted from the social media platform in 2009 for research purposes. Previously, some accounts appeared highly suspicious but their tweets had not been removed and they were not suspended. FRAUDAR was now able to find more than 4,000 fraudulent accounts.
As the researchers explain, they’re not finding anything criminal here, but these sorts of frauds can really undermine people’s faith in online reviews and behaviors.
The new algorithm is fast and can see through camouflage that fraudsters use to make themselves look legitimate. It’s also available as open-source code and it doesn’t require researchers to target anybody.
The researchers hope that by making it open-source, social media platforms could become safer.
Source: Carnegie Mellon University via ScienceDaily (https://www.sciencedaily.com/releases/2016/09/160908083826.htm)