Open source algorithm helps spot social media shams

Carnegie Mellon researchers say algorithm can see through facades fraudsters use to make themselves look genuine

open-source-algorithm-helps-spot-social-media-shams

Researchers from Carnegie Mellon University say they have developed an open source algorithm that can help spot social media frauds trying to sway valuable community influence.

+More on Network World: Star Trek turns 50!+

“Given the rise in popularity of social networks and other web services in recent years, fraudsters have strong incentives to manipulate these services. On several shady websites, anyone can buy fake Facebook page-likes or Twitter followers by the thousands. Yelp, Amazon and TripAdvisor fake reviews are also available for sale, misleading consumers about restaurants, hotels, and other services and products. Detecting and neutralizing these actions is important for companies and consumers alike,” the researchers wrote in a paper outlining their algorithm known as FRAUDAR.

According to Carnegie Mellon researchers the new algorithm makes it possible to see through camouflage fraudsters use to make themselves look legitimate.

According to Christos Faloutsos, professor of machine learning and computer at Carnegie Mellon the state-of-the-art for detecting fraudsters, with tools such as NetProbe, is to find a pattern known as a “bipartite core.” These are groups of users who have many transactions with members of a second group, but no transactions with each other. This suggests a group of fraudsters, whose only purpose is to inflate the reputations of others by following them, by having fake interactions with them, or by posting flattering or unflattering reviews of products and businesses, he said in a statement.

But fraudsters have learned to camouflage themselves. They link their fraudulent accounts with popular sites or celebrities, or they use legitimate user accounts they have hijacked. In either case, they try to look “normal,” Faloutsos said.

FRAUDAR can prune away this camouflage Faloutsos said. “Essentially, the algorithm begins by finding accounts that it can confidently identify as legitimate — accounts that may follow a few random people, those that post only an occasional review and those that otherwise have normal behaviors. This pruning occurs repeatedly and rapidly. As these legitimate accounts are eliminated, so is the camouflage the fraudsters rely upon. This makes bipartite cores easier to spot.”

Check out these other hot stories:

“Guccifer” gets 52 months in prison for hacking crimes

Researchers sport system to pull rare earth materials from used hard drives

1,650lb 3D printed aircraft tool sets Guinness World Record

Cisco buys into containers with Container X acquisition

8 cool “habitable” planets

Cisco: US Trade Rep. backing Arista product import ban

FBI: Bank robbery? There’s an app for that

Much ado about the ransomware scourge

US senators want airline IT meltdowns to end

Prize competitions for tough IT, high-tech problems all the rage

Cisco to jettison 5,500 jobs, will reinvest in cloud, IoT & more

Cisco uncovers security threat in industrial control system

Open vSwitch finds new home at the Linux Foundation

What will space living look like? NASA picks 6 habitat prototypes

Branch office links, big bandwidth needs drive SD-WAN evolution

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.
Must read: Hidden Cause of Slow Internet and how to fix it
Notice to our Readers
We're now using social media to take your comments and feedback. Learn more about this here.