Started By
Message
locked post

Data Scientists Can Use Your Anonymous Data To ID You

Posted on 7/23/19 at 11:29 am
Posted by StupidBinder
Jawja
Member since Oct 2017
6392 posts
Posted on 7/23/19 at 11:29 am
Careful what you share out here on the interwebs...

It’s getting easier and easier to cobble together people’s identity from the little random “anonymized” bits of info floating around.

quote:

But new research indicates that even when data is stripped of any identifiable factors, it doesn’t require a lot of mental gymnastics to piece together certain information and figure out, with pretty high confidence, who the “anonymous” user in the dataset is.

In other words, anonymized data is not so anonymous.


quote:

Using a machine learning model and datasets that included up to 15 identifiable characteristics—such as age, gender, and marital status—the researchers were able to accurately reidentify 99.98 percent of Americans in an anonymized dataset, according to the study. For their analyses, the researchers used 210 different data sets that were gathered from five sources including the U.S. government that featured information on more than 11 million individuals.


Stay woke
This post was edited on 7/23/19 at 11:32 am
Posted by Cump11b
Member since Sep 2018
2026 posts
Posted on 7/23/19 at 11:32 am to
quote:

Anonymous Data is Not So Anonymous


Anyone who thinks "anonymous"employer surveys are Anonymous are kidding themselves
Posted by aTmTexas Dillo
East Texas Lake
Member since Sep 2018
24643 posts
Posted on 7/23/19 at 11:32 am to
quote:

Careful what you share out here on the interwebs...


Yeah, the interweb is so fun and we can do and see so much. And it's all free. Imagine that. There is not any catch in it is there?
Posted by GreatLakesTiger24
Member since May 2012
61518 posts
Posted on 7/23/19 at 11:33 am to
Yikes
Posted by Langland
Trumplandia
Member since Apr 2014
15382 posts
Posted on 7/23/19 at 11:34 am to
We're are about 10 years away from the government knowing exactly where you are and what you are doing at any time.
This post was edited on 7/23/19 at 11:34 am
Posted by ExtraGravy
Member since Nov 2018
978 posts
Posted on 7/23/19 at 11:34 am to
(no message)
This post was edited on 11/24/20 at 9:57 pm
Posted by YouAre8Up
in a house
Member since Mar 2011
12792 posts
Posted on 7/23/19 at 11:35 am to
quote:

We're are about 10 years away from the government knowing exactly where you are and what you are doing at any time.


We are already there. They know what you are doing.
Posted by magildachunks
Member since Oct 2006
36631 posts
Posted on 7/23/19 at 11:36 am to
quote:

And it's all free.



How you getting free internet?
Posted by StupidBinder
Jawja
Member since Oct 2017
6392 posts
Posted on 7/23/19 at 11:36 am to
quote:

Yeah, the interweb is so fun and we can do and see so much. And it's all free. Imagine that. There is not any catch in it is there?


I never put anything in those things that I wouldn’t be willing to tell my boss or his boss to their faces.
Posted by aTmTexas Dillo
East Texas Lake
Member since Sep 2018
24643 posts
Posted on 7/23/19 at 11:39 am to
quote:

How you getting free internet?


You think too deeply. Lighten up on the brain activity.
Posted by Oddibe
Close to some, further from others
Member since Sep 2015
6756 posts
Posted on 7/23/19 at 11:44 am to
quote:

are about 10 years away from the government knowing exactly where you are and what you are doing at any time.
Your credit card company can pretty much do that now based on historical spending patterns. They know how often you eat out and where you regularly dine, shop buy gas and many other things that give give them the ability project your future actions.
Posted by StupidBinder
Jawja
Member since Oct 2017
6392 posts
Posted on 7/23/19 at 11:46 am to
quote:

Yeah, the interweb is so fun and we can do and see so much. And it's all free. Imagine that. There is not any catch in it is there?


In fairness, the “agreement” has always been along the lines of “I’ll give you some info about me in exchange for your free email, search engine and the like...as long as you keep it private”.

I don’t think that’s a unreasonable expectation. I work in financial services and we have to go to pretty great lengths to protect information. So it can be done properly, but what this article shows it that there are a lot of entities who seem to think that anonymizing data simply means hiding the name and SSN, and that won’t cut it if we all have an interest in maintaining the “free internet” model.
Posted by slackster
Houston
Member since Mar 2009
91871 posts
Posted on 7/23/19 at 11:49 am to
quote:

the researchers were able to accurately reidentify 99.98 percent of Americans in an anonymized dataset


How do they know?
Posted by StupidBinder
Jawja
Member since Oct 2017
6392 posts
Posted on 7/23/19 at 12:07 pm to
quote:

How do they know?


It appears that at least some of the data was purchased from data brokers who are obligated by law to anonymize the data they sell. The researchers told the broker what they wanted to do and the brokers verified the results.
Posted by slackster
Houston
Member since Mar 2009
91871 posts
Posted on 7/23/19 at 12:20 pm to
quote:

It appears that at least some of the data was purchased from data brokers who are obligated by law to anonymize the data they sell. The researchers told the broker what they wanted to do and the brokers verified the results.


It seems like telling them they were right on 99%, even if they didn't tell them individual results, would be a breach of privacy.
Posted by Hammertime
Will trade dowsing rod for titties
Member since Jan 2012
43031 posts
Posted on 7/23/19 at 12:27 pm to
You honestly think any companies care about your online privacy? Everything you do is sold ten times over. Hell, everything you do all day long is sold to whoever wants to know. Why collect data if you can't sell it?
Posted by slackster
Houston
Member since Mar 2009
91871 posts
Posted on 7/23/19 at 12:36 pm to
quote:

You honestly think any companies care about your online privacy? Everything you do is sold ten times over. Hell, everything you do all day long is sold to whoever wants to know. Why collect data if you can't sell it?


I understand they sell it. I'm asking how you can justify telling researchers they got 99% of the people correct. That sort of validation isn't much different than just giving them the names.
Posted by DarthRebel
Tier Five is Alive
Member since Feb 2013
26120 posts
Posted on 7/23/19 at 1:03 pm to
quote:

How you getting free internet?


Free internet is everywhere. Seriously it is everywhere.
Posted by DarthRebel
Tier Five is Alive
Member since Feb 2013
26120 posts
Posted on 7/23/19 at 1:06 pm to
quote:

Using a machine learning model and datasets that included up to 15 identifiable characteristics—such as age, gender, and marital status—the researchers were able to accurately reidentify 99.98 percent of Americans in an anonymized dataset,


15 is a lot of traits to use. Most humans would be able to do the same, although it would take longer.
Posted by Redbone
between St. Amant and Port Vincent
Member since Sep 2012
20738 posts
Posted on 7/23/19 at 1:51 pm to
quote:

Data Scientists Can Use Your Anonymous Data To ID You
Imagine that !!
first pageprev pagePage 1 of 2Next pagelast page

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on X, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

FacebookXInstagram