The team found that scripts used by scammers contain multiple paths, which can be simplified into four different stages: “Our artificial intelligence model looks for particular phrases and keywords, and common transitions between topics that might reflect steps in a scam script,” he says, adding that the 100 hours of transcript is a rich data source, as just one hour of audio can include 5000 to 9000 words. Team member Dr Ian Wood is an expert in natural language processing – which he admits is “somewhat a dark art". Research offers hope for MND and Alzheimer's gene treatment.Selected samples were subject to sentiment analysis, and machine-learning models were used to find patterns in the calls. Kaafar's team transcribed more than 300 scam calls published on YouTube, most recorded by people who pretend to fall for a scam, then eventually reveal their ploy and chastise the scammer. “Ours is one of the only studies to unpack the content of scam calls and the psychological tricks used by attackers in depth,” he says. Harder to address are the social engineering techniques that scammers use to manipulate their victims into revealing personal and account details, purchasing online vouchers or transferring funds to bank accounts controlled by the scammer. Kaafar says that while a range of technology fixes have been developed such as blocking known bad numbers and using pattern recognition on outgoing calls, these capture only a small portion of the vast influx of scam calls. Technology can only be part of the solutionĭespite new rules introduced by the Australian government allowing telcos to block more than 200 million scam calls during 2021, millions more scam calls still get through each week, Kaafar says.īusted: The team led by Professor Kaafar (pictured) used machine-learning techniques and natural language processing to uncover scam ‘scripts’. “Advances in technology allow attackers to hide their identities and use things like recorded robocalls and VOIP platforms like Skype to reduce call costs, and lower their own risk,” he says. “I have even known a post-doctoral researcher, with years of experience in cybersecurity, who was tricked out of $8000 in a phone scam,” he adds.ĭuring 2021 alone, phone scammers stole more than $100 million from Australians via more than 144,000 incidents reported to the ACCC’s ‘Scamwatch’ service – and that’s just the tip of a fast-growing iceberg, according to Kaafar.
Ours is one of the only studies to unpack the content of scam calls and the psychological tricks used by attackers in depth.
“Most people have either been targeted themselves, or know someone who has fallen victim to scammers, because it’s so common, so relentless and in many ways, so clever,” says Kaafar. These findings will help develop better ways to detect and prevent scams which account for the human element which is critical to scammers’ success. The team, headed by Professor Dali Kaafar, used machine-learning techniques and natural language processing to uncover scam ‘scripts’ that use various topics and emotions.
A team of researchers from Macquarie University’s Cyber Security Hub has analysed the content of more than 100 hours of scam phone calls to identify clear call ‘stages’ and pinpoint the social engineering techniques scammers use on their victims.Ĭon job: During 2021 alone, phone scammers stole more than $100 million from Australians.
0 kommentar(er)
