Researchers Can ID Anonymous Twitterers
narramissic writes "In a paper set to be delivered at an upcoming security conference, University of Texas at Austin researchers showed how they were able to identify people who were on public social networks such as Twitter and Flickr by mapping out the connections surrounding their network of friends. From the ITworld article: 'Web site operators often share data about users with partners and advertisers after stripping it of any personally identifiable information such as names, addresses or birth dates. Arvind Narayanan and fellow researcher Vitaly Shmatikov found that by analyzing these 'anonymized' data sets, they could identify Flickr users who were also on Twitter about two-thirds of the time, depending on how much information they have to work with.'"
Who ever promised this data would be anonymous? Do you really expect privacy when posting personal stuff on line, even if you don't sign your name in advance?
John
Slashdotters care about privacy. People on these social networking sites want their lives to be on show for everyone. I don't think people who twit every 5 minutes where they are and what they are doing are really to concerned about their privacy.
Support a great indie game: http://www.abaddon360.com
Are there really any surprises here? Social networks behave a lot like the Internet, with many routes pointing to your front door.
For example, use whatever falese names you want. Your email address makes a dandy primary key squirreled away in all your friends mailboxes, just waiting for Facebook to Hoover it up and join the dots.
Your privacy and anonymity is defined by the aggregate social stupidity of your friends.
Xix.
"Everything is adjustable, provided you have the right tools"
We have an FAQ about this paper. It answers many of the misconceptions expressed in the comments here. In particular, our algorithm applies to much more than public social networks like twitter and flickr. A variety of networks including the phone call network are being shared behind your back in anonymous form, and our de-anonymization techniques apply just as much. You'll probably agree that people expect more privacy there. See my blog for a variety of demonstrations and thought-experiments of de-anonymization.
http://www.guardian.co.uk/technology/2009/mar/26/seth-finkelstein-google-advertising
"Google recently took another step along the path of surveillance as a service, launching what it called "interest-based advertising", and which everyone else calls "behavioural targeting". These are systems that collect extensive personal data, for marketing purposes. To best understand the issues,"
http://sethf.com/infothought/blog/archives/001422.html
I once upon a time worked for a statistics agency and even without names and addresses it is surprisingly easy to identify people in anonymous data, even anonymised unit record data can be deconstructed to some degree. Depending on what you want to achieve don't even need to identify them.
Marrying up these datasets and ideas would be gruesome.
The Singularity is closer than you think
Quant