SoylentNews is people
SoylentNews
On Wednesday, the Federal Trade Commission (FTC) announced another contest to design a system to "identify unwanted robocalls received on landlines or mobile phones, and block and forward those calls to a honeypot." The agency will select "up to five contestants" as part of what it’s calling "Robocalls: Humanity Strikes Back."
The first qualifying phase launches Wednesday and runs through June 15, 2015 at 10:00pm Eastern Time, while the final phase concludes at DEF CON 23 on August 9, 2015.
Here's the FTC contest page. There's another similar contest (with no cash prize) being held "as part of the National Day of Civic Hacking." It appears they have done something similar in previous years as well.
This discussion has been archived.
No new comments can be posted.
FTC to Award $25,000 for the Best Honeypot Design to Trap Robocalls
|
Log In/Create an Account
| Top
| 26 comments
| Search Discussion
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Live in a world of your own, but always welcome visitors.
(Score: 3, Interesting) by kaszz on Thursday March 05 2015, @02:57PM
Just listening into the SS7 signaling ought to reveal these spam callers. A small pool of originating numbers that calls a large pool of numbers within a short time. Just identify that source pool and block them?
The technology is there. It's just ineptitude withing telecommunication companies that hinders this from happening. But then handling lot's of calls means money so why "fix it".
(Score: 4, Insightful) by MrGuy on Thursday March 05 2015, @03:05PM
Except that this also describes outgoing call centers, which have a variety of legitimate purposes. For example, a customer service team that contacts customers with problem orders placed online would look similar (small number of outgoing numbers, large number of contacted numbers, frequent calls, many short duration calls because a lot of the time you'll get voicemail....)
Not saying it's not possible to identify "highly likely to be robocalling" by data mining, just that there's a difference between "a lot of outbound calls" and "robocalls"
(Score: 3, Interesting) by kaszz on Thursday March 05 2015, @03:15PM
Add call length to the data mining? Lack of anyone calling back etc..
(Score: 3, Informative) by Thexalon on Thursday March 05 2015, @03:32PM
It's not a small pool of originating numbers, because the robocallers spoof the originating numbers to get around the fact that there are blacklisting services out there for phone numbers.
And as a sibling poster points out, even if they didn't do that a legitimate call center (e.g. a utility customer service department) would look identical to a robocaller from the metadata-only viewpoint.
"Think of how stupid the average person is. Then realize half of 'em are stupider than that." - George Carlin
(Score: 1, Interesting) by Anonymous Coward on Thursday March 05 2015, @06:02PM
as a sibling poster points out, even if they didn't do that a legitimate call center (e.g. a utility customer service department) would look identical to a robocaller from the metadata-only viewpoint.
Furthermore, we should expect that whatever differences in meta-data there might be would be quickly emulated by the spammers. No incoming calls? Start sending in bogus incoming calls. Duration too short? Start making long bogus calls to phone numbers under control of the spammers.
These guys may be assholes, but to assume they aren't clever is just wishful thinking.
(Score: 2) by kaszz on Friday March 06 2015, @08:23AM
There's a difference in SS7 numbers and the one presented as caller-identification to the customer.
One could perhaps use voice fingerprinting used in several successive calls as well.