Article picture is an IBM iDataPlex system.
Algorithm can pick out almost any American in supposedly anonymized databases
81–90 of 101 posts
Re: Algorithm can pick out almost any American in supposedly anonymized databases
#82I'm a programmer in the GP data analysis world. We use the term 'pseudonymization' for this kind of data. 'Anonymization' is used solely to refer to, say, 'the sum total of diabetes patients this practice has' (that would be anonymous patient data; it would not be anonymous relative to the GP office this refers to): Aggregated data that can no longer be reduced to a single individual at all. The term raises questions…
We covered this in a DB course with the term k-anonymity which seems to be standard in the literature, where a dataset is k-anonymous if every combination of characteristics (that can identify users) has at least K users. So in your case that dataset has only the 1-anonymity property, but you can set a k>1 and change the data set to satisfy it and improve the anonymity. Eg. if age was just stored as 90+ and there's a…
>I guess then the interesting question is how high does k have to be to call it anonymous vs pseudonymous.
I think that for any size k less than the total size of the database, it is not anonymous. In cases like this, an overly strict definition favoring privacy is the only way to protect people. Similar to how we call 17 year olds children and treat them as such under law even though a 17 year old is far closer to an 18 year old than they are to a 5 year old (yes, there are some exceptions, but these are all explicitly called out). Another example of such an extreme is concerning falsifying data or making false statements. Even a single such statement, regardless of the number of true statements, destroys credibility once found when trust is extremely important. This is why even a single such statement can get one found in contempt of court or destroy a scientist's entire career (and even cast doubt on peers who were innocent).
Overall it is quite messy because it is a mix of a technical problem with a people problem.
Re: Algorithm can pick out almost any American in supposedly anonymized databases
#83In any database whose fields have at least 4 meaningful ranges, 16 fields give 4.000.000.000 possibilities. Now I am on my iphone. As long as ranges are meaningful (i.e. they do divide the group in somewhat even parts), the individuating possibilities are HUGE. And fields do have usually many more than 4 ranges. Anonymizing datasets is a weasel term. The database is secure or it is not. As any database is quite likel…
An anonymized version would be one where you have a tally for each value/category in each field, without any correlating table between the fields. Only store the histograms.
Re: Algorithm can pick out almost any American in supposedly anonymized databases
#84The great contribution of Differential Privacy theory is to quantify just how little use you can get out of aggregate data before individuals become identifiable. Unfortunately, Differential Privacy proofs can be used to justify applications which turn out to leak privacy when the proofs are shown to be incorrect after the fact, when the data is already out there and the damage already done. Nevertheless, it is instr…
The part I can't wrap my head around is how to mitigate (future proof) unforeseen leakage and correlations. The example I keep going back to is deanonymizing movie reviews (chronologically correlating movie rentals and reviews). And, frankly, I'm just not clever enough to imagine most attacks. If nothing else, I appreciate the Differential Privacy effort, if only to show the problem space is wicked hard. I worked in…
> The part I can't wrap my head around [...] The example I keep going back to is deanonymizing movie reviews
The reason is that you are thinking of an example that's not nicely compatible with differential privacy. The basic examples of DP would be something like a statistical query: approximately how many people gave Movie X three stars? You can ask a bunch of those queries, adding some noise, and be protected against re-identification.
You can still try to release a noisy version of the whole database using DP, but it will be very noisy. A basic algorithm (not good) would be something like
For each entry (person, movie):
with probability 0.02, keep the original rating
otherwise, pick a rating at random
Re: Algorithm can pick out almost any American in supposedly anonymized databases
#85In any database whose fields have at least 4 meaningful ranges, 16 fields give 4.000.000.000 possibilities. Now I am on my iphone. As long as ranges are meaningful (i.e. they do divide the group in somewhat even parts), the individuating possibilities are HUGE. And fields do have usually many more than 4 ranges. Anonymizing datasets is a weasel term. The database is secure or it is not. As any database is quite likel…
An anonymized version would be one where you have a tally for each value/category in each field, without any correlating table between the fields. Only store the histograms.
Well, yes, but then that is not what advertisers want... That is the thing.
Whenever a large enough database exists with individual data, we are doomed.
Re: Algorithm can pick out almost any American in supposedly anonymized databases
#86Would differential privacy fix this problem? I heard that new US census will use it.
At ICML 2019 there was a good keynote by the chief scientist of the Census Bureau. The 2010 census was before a good understanding of differential privacy, so it wasn't really done correctly. The census bureau actually went back and bought a bunch of commercially available data, and deanonymized ~70% of the individuals in the US (if I remember correctly). So they wanted to do it better this time, and they seem to be taking the state of the art into account. It was much more impressive than I was expecting, actually.
https://icml.cc/Conferences/2019/ScheduleMultitrack?event=43... there is a video link on this page
Re: Algorithm can pick out almost any American in supposedly anonymized databases
#87I'm a programmer in the GP data analysis world. We use the term 'pseudonymization' for this kind of data. 'Anonymization' is used solely to refer to, say, 'the sum total of diabetes patients this practice has' (that would be anonymous patient data; it would not be anonymous relative to the GP office this refers to): Aggregated data that can no longer be reduced to a single individual at all. The term raises questions…
>> Aggregated data that can no longer be reduced to a single individual at all.
Even aggregated data will loose the anonymisation characteristics when we are speaking of low volumes of data.
Number of cancer patients in the area A: 1 Number of residents in the area A: 1
Re: Algorithm can pick out almost any American in supposedly anonymized databases
#88I'm a programmer in the GP data analysis world. We use the term 'pseudonymization' for this kind of data. 'Anonymization' is used solely to refer to, say, 'the sum total of diabetes patients this practice has' (that would be anonymous patient data; it would not be anonymous relative to the GP office this refers to): Aggregated data that can no longer be reduced to a single individual at all. The term raises questions…
We covered this in a DB course with the term k-anonymity which seems to be standard in the literature, where a dataset is k-anonymous if every combination of characteristics (that can identify users) has at least K users. So in your case that dataset has only the 1-anonymity property, but you can set a k>1 and change the data set to satisfy it and improve the anonymity. Eg. if age was just stored as 90+ and there's a…
Isn't part of that problem how you define a "characteristic" (and thus, the combinations of them)? Because if it has k-anonymity for a certain set of characteristics, but you invent a new characteristic (based from combining info, datamining and perhaps other sources), the property doesn't necessarily hold any more.
Re: Algorithm can pick out almost any American in supposedly anonymized databases
#89Earlier quoted context omitted.
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.
Can you provide a link to your paper?
The article is available here, in open access: https://www.nature.com/articles/s41467-019-10933-3
Re: Algorithm can pick out almost any American in supposedly anonymized databases
#90Effective de-identification is much harder than people think.