Algorithm Red Flags Potentially Dangerous Domain Names at Time of Purchase
By Sara Castellanos Oct 31, 2016 9:51 am ET A machine-learning algorithm that can detect malicious domain names as soon as they’re purchased has peaked the interest of at least one corporate IT executive. Invented by five researchers at Princeton University and the University of California, the code works by scanning for 22 “features,” or red flags that are consistent with suspicious behavior. They include names that are registered in bulk by the hundreds, names that are variations of the same name, random-looking names, and names with numerical characters. The tool would be helpful if it were used by the hundreds of registrars offering domain names for purchase in order to stop cyber criminals planning attacks as they’re purchasing the names, said Vincent Weafer, senior vice president at McAfee Labs…
Link to Full Article: Algorithm Red Flags Potentially Dangerous Domain Names at Time of Purchase
Sharing
