So today its mostly reputation scoring aggregators.
The two biggest that come to mind are whitepages.com (they own Hiya) and
NextCaller.
There is some secret sauce, but the scoring algorithm is based on
incoming query volume, LERG lookups, LIDB lookups, and of course user
feedback from the freemium app Hiya which is in wide distribution.
I know nextcaller has some deeper hooks with major providers allowing
them more accurate info (such as correlating an active call on the
owning provider with the call youre dipping about etc), but their focus
isnt SPIT protection, its an enterprise focused product and is priced as
such.
Google (Android) now ships with this in the OS, and I presume they are
leveraging one of these datasources since the call classifications match
my whitepages dips. For network providers doing this, its generally just
an appserver the call passes through that will do the dip, and modify
the call headers for later behavioral changes (or at least that's how I
have built it and seen it implemented). In a freeswitch ecosystem I just
build the dialplan conditionally based upon the dip (so changing PAI or
potentially even punting to VM).
I have reached out to my contact with whitepages.com about how to
contest an inaccurate ranking in the past, and they dont see well
equipped to do much about it yet, BUT in my case it was a newly assigned
number that had been previously used by a bad actor and abandoned. It
righted itself pretty quick when the bad call volumes stopped.
On 8/30/2018 3:27 PM, Alex Balashov wrote:
Hi,
So, of course, it is a known problem in the legitimate side of the lead
engagement call centre, survey, etc. business that ANIs get flagged as
"telemarketer" pretty quickly and start showing up that way on people's
phones.
Your normative viewpoint on that may vary depending on what you think of
lead gen calls, but regardless, a reasonable person would differentiate
between:
(1) Legitimate outbound dialing operations that are following up with
leads who provided their phone number and agreed to be called (at least,
as a matter of clicking "accept" on _something_), or for some other
legitimate mass-dialing purpose, and who actually own the DID inventory
from which they present local-market ANIs and can in fact be reached on
those numbers;
(2) Illegal spammers who use fake ANIs and call people who did not in
any way consent to being called by them.
I am trying to learn a bit more about how this is done and what a
legitimate, above-board business can do about it.
Ignoring the factor of third-party call-screening apps (which most
people with a mobile don't use), where is this generally implemented? As
I understand it, T-Mobile do it on the network level. I have T-Mobile
myself, and probably 2/3rds of unfamiliar numbers, including quite
legitimate ones, show up as "Scam Likely" — I know that's come up on the
list before. AT&T displays "Telemarketer"; do they do it that way too,
or do they use a Google Android feature for that which they enable as
part of their carrier defaults for carrier-issued phones? What about
other carriers and Android?
As far as I know, Apple don't do anything like this. Do people with
iPhones just not receive this "service"? How does that work?
Asking where the central, or the most influential authority lies and who
provides it goes to the heart of the real question, which is: what can a
legitimate business do if their number has been blacklisted this way? As
I understand it, the maintainers of these lists, along with the criteria
for getting on them, are elusive and inscrutable, and there's really no
recourse and no appeals process. I furthermore understand that this has
led to the widespread approach of rotating ANIs, but that's a losing
battle; they get flagged too. I imagine it won't be long before the
criteria for "Scam Likely" are just "number appears to call lots of
numbers in this rate centre and otherwise hasn't been around very long".
But this is all just conjecture on my part; I really don't know much
about how my carrier, anyone's carrier, or some BigCo that's behind my
mobile OS decides that a call is a "telemarketer" or "scam" call. If
anyone can shed some light on how this really works and what, if
anything can be done about it, I would be most appreciative.
_______________________________________________
VoiceOps mailing list
[email protected]
https://puck.nether.net/mailman/listinfo/voiceops
