Researchers from North Carolina State University revealed on Wednesday a new method to foil malicious robocalls.
SnorCall uses artificial intelligent to analyze robocalls. It can, in many cases, give law enforcement agencies and other stakeholders the information needed to track down bad actors.
It is not a paper Researchers explained at Usenix Security Symposium held in Boston how SnorCall had been used to record 232 723 robocalls on over 60,000 phone numbers provided by Bandwidth. This telecommunications provider was providing them with the project.
SnorCall got its name from a machine learning network called Snorkel. The robocalls are then transcribed by the network and analyzed. Snorkel is a machine-learning network that allows researchers to manage and build training datasets, without having to manually label them. This reduces the amount of time spent on manual labeling from weeks or months down into a matter of hours or days.
Sathvik Prasad is a Ph.D. Student at NC State who was the first author on the paper.
He told TechNewsWorld that “our framework is flexible enough for us to categorize all types of robocalls.” “In our paper we show how to do this for social security and technical support scams which are predominantly malicious.”
He added, “There are a lot of benign but annoying telemarketing callers like those for auto warranties.” “We also collected them, but they are not as harmful as border patrol or impersonation calls from social security. The most harmful calls to society were those that we studied in depth.
Finding Robocallers
SnorCall uses labels created by Snorkel in order to identify the call. Is it a mention of a particular company or government program Is it asking for specific information about you? Does it ask for specific personal information? Is it asking for money? Does it ask for money?
All of this information is then fed into a database which can be used to identify patterns or behaviors. Researchers also showed how to find phone numbers which could be traced directly back to those who made the robocalls.
Brad Reaves is an assistant professor at NC State and the corresponding author of a paper that describes the research. TechXPlore.
“And about 45% of the robocalls we analyzed did include this ‘callback number’ strategy,” he continued. SnorCall’s ability to extract these callback number gives regulators and law enforcers something with which to work. They can identify which phone service provider issued these numbers, and then who opened those account.
“What we did — extracting callback numbers from robocalls — had never been done before,” added Prasad. It also allows us the ability to connect those numbers to global calling infrastructure.
He noted that “spoofing” caller ID numbers was an almost universal tactic used by malicious robocalls — with 95% of the calls spoofing caller IDs.
Out of Control Problem
“Robocalls have gotten out of hand in the U.S.,” said Maria-Kristina Hayden. CEO and founder, Outfoxm, is a company that specializes in cyber hygiene and resilience.
She told TechNewsWorld that some estimates suggest over 50 billion phone calls will be made to U.S. customers in 2022. “I hear from hundreds of individuals a month who are fed up with the cadence of ‘scam likely’ calls to their cell and home phones.”
Liz Miller, vice-president and principal analyst at Stratfor, said that the robocall issue is complex. Constellation ResearchCupertino is home to, an advisory and research firm for technology.
She told TechNewsWorld that scam calls are on the rise, and they often target disproportionately vulnerable groups, like elderly non-English speaking people who are afraid their social security numbers will be used in criminal acts.
“But in reality, more businesses, government offices, and organizations are using these automated recorded messages to quickly distribute information. “A weekly recorded message sent by a principal to parents each week is an example,” she added.
She said that “Robocalls do not have to be bad or unwanted.” “However there is a patchy ability to separate the good from bad.” YouMail, the company that released these numbers, noted in January 2022 that, on average, 1,831 robocalls were made per second.
Believable deceptions
Technologist Jeff Kagan Many of these calls can pose a problem for consumers. “Robocalls can be the beginning of a theft process,” he said to TechNewsWorld.
Hayden said that robocalls were a problem, because they were malicious and became more convincing, trickier, and annoying.
“Even the national Do Not Call list and paid apps that are supposed to prevent calls are not very effective — and I’ve tried many of them,” she said.
Fraudsters are increasingly using the global context of their scams to make them more convincing. She continued: “For example, during Covid, scammers called pretending to be pharmacies, calling with Covid results or vaccine appointment.”
“Scammers also target specific generations, using topics that could lure them in. For example, posing as Medicare and calling elderly populations,” she said.
Miller asserted that some robocalls could be frightening. She said that if the caller does not know it is a scam, they can find it downright frightening to be told their passport will be held by a consulate until a fine is paid.
She also noted that scammers using robocalls are becoming more sophisticated. She mentioned a scam involving a tag-along scam that was tied to Camp Lejeune litigation.
While legitimate calls were made to join an ongoing legal action, robocalls told people to call a particular number where they would be asked to provide a credit card number in order to join the case.
Best To Be Skeptical
Kagan says that consumers should be skeptical when it comes to robocalls. You should know that most of the time, if you receive a call, an email or text from someone you do not know, it is a scam. Ignore it.
He said that he received dozens of calls, voicemails and emails every day. There’s no way to filter them all.
He said, “It is unfortunate that we have come to this.” “But, we have.”
