zlacker

[return to "Algorithm can pick out almost any American in supposedly anonymized databases"]
1. shusso+O4[view] [source] 2019-07-24 10:39:25
>>zoobab+(OP)
The article title misses a bit of nuance from the paper which is specifically talking about re-identification.

e.g from the paper:

"We show that, as a male born on July 31, 1945 and living in Cambridge (02138), the information used by Latanya Sweeney at the time, William Weld was unique with a 58% likelihood (ξx = 0.58 and κx = 0.77), meaning that Latanya Sweeney’s re-identification had 77% chances of being correct. We show that, if his medical records had included number of children—5 for William Weld—, her re-identification would have had 99.8% chances of being correct!"

◧◩
2. Freak_+87[view] [source] 2019-07-24 11:10:32
>>shusso+O4
What accounts for that remaining 2‰ of uncertainty?
◧◩◪
3. Cynddl+ke[view] [source] 2019-07-24 12:19:16
>>Freak_+87
Co-author here. We designed a statistical model, which is never 100% sure a re-identification is correct. There is, e.g., a non-null probability that two individuals in the US share 5, 10, or even 15 demographics attribute.
◧◩◪◨
4. mnky98+ms[view] [source] 2019-07-24 13:56:14
>>Cynddl+ke
Can you provide a link to your paper?
◧◩◪◨⬒
5. Cynddl+SK[view] [source] 2019-07-24 15:49:27
>>mnky98+ms
The article is available here, in open access: https://www.nature.com/articles/s41467-019-10933-3
[go to top]