http://33bits.org/2008/11/12/57/ The End of Anonymous Data Tue, 09 Mar 2010 07:37:18 +0000 http://wordpress.com/ hourly 1 http://33bits.org/2008/11/12/57/#comment-1043 Cookies, Supercookies and Ubercookies: Stealing the Identity of Web Visitors « 33 Bits of Entropy Thu, 18 Feb 2010 07:50:16 +0000 http://33bits.org/2008/11/12/57/#comment-1043 [...] Group affiliations, just like your movie-watching history and many other types of attributes, are sufficient to fingerprint a user. There’s a high chance there’s no one else who belongs to the same set of groups that you do (or is even close). [Aside: I used this fact to show that Lending Club data can be de-anonymized.] [...] [...] Group affiliations, just like your movie-watching history and many other types of attributes, are sufficient to fingerprint a user. There’s a high chance there’s no one else who belongs to the same set of groups that you do (or is even close). [Aside: I used this fact to show that Lending Club data can be de-anonymized.] [...]

]]> http://33bits.org/2008/11/12/57/#comment-675 CA Mon, 14 Sep 2009 02:45:31 +0000 http://33bits.org/2008/11/12/57/#comment-675 Extracting zip codes is rather easy, the data is available through the census beauru. It's like stealing candy from a baby :P. I had to do it for the mailers for my new not-for-profit website, in just california. Extracting zip codes is rather easy, the data is available through the census beauru. It’s like stealing candy from a baby . I had to do it for the mailers for my new not-for-profit website, in just california.

]]> http://33bits.org/2008/11/12/57/#comment-251 Gary in Clearwateer Thu, 16 Apr 2009 19:16:56 +0000 http://33bits.org/2008/11/12/57/#comment-251 Conceptually it makes sense, but as most of the posters have already acknowledged, I would think that the cumulative distribution of information would be a significant target for data mining. And they identify particularly 'sensitive' areas as financial data and loan description, but what about other types of data that they may consider 'not-so-sensitive' but which could ultimately compromise someone's identity when collected longitudinally? Conceptually it makes sense, but as most of the posters have already acknowledged, I would think that the cumulative distribution of information would be a significant target for data mining. And they identify particularly ’sensitive’ areas as financial data and loan description, but what about other types of data that they may consider ‘not-so-sensitive’ but which could ultimately compromise someone’s identity when collected longitudinally?

]]> http://33bits.org/2008/11/12/57/#comment-240 Lending Club’s CEO Explains Loan Funding, Privacy Issues | P2P Lending News Thu, 09 Apr 2009 16:35:43 +0000 http://33bits.org/2008/11/12/57/#comment-240 [...] have probably seen this post about how Lending Club’s practice of publishing certain borrower information makes the [...] [...] have probably seen this post about how Lending Club’s practice of publishing certain borrower information makes the [...]

]]> http://33bits.org/2008/11/12/57/#comment-221 fsu.edu | Artech Blog Mon, 30 Mar 2009 15:19:45 +0000 http://33bits.org/2008/11/12/57/#comment-221 [...] remind us that websites that collect and republish seemingly innocuous facts about their users are often vulnerable to data mining. It doesn’t matter whether you keep the users’ names and addresses secret — the facts [...] [...] remind us that websites that collect and republish seemingly innocuous facts about their users are often vulnerable to data mining. It doesn’t matter whether you keep the users’ names and addresses secret — the facts [...]

]]> http://33bits.org/2008/11/12/57/#comment-160 News » Last.fm and the Diabolical Power of Data Mining Fri, 06 Mar 2009 21:16:59 +0000 http://33bits.org/2008/11/12/57/#comment-160 [...] remind us that websites that collect and republish seemingly innocuous facts about their users are often vulnerable to data mining. It doesn’t matter whether you keep the users’ names and addresses secret — the [...] [...] remind us that websites that collect and republish seemingly innocuous facts about their users are often vulnerable to data mining. It doesn’t matter whether you keep the users’ names and addresses secret — the [...]

]]> http://33bits.org/2008/11/12/57/#comment-28 Arvind Thu, 27 Nov 2008 04:29:19 +0000 http://33bits.org/2008/11/12/57/#comment-28 Fred: very interesting. I wonder if there is a way to extract the zip codes automatically. I'm guessing it should be possible with a bit of javascript-fu. Fred: very interesting. I wonder if there is a way to extract the zip codes automatically. I’m guessing it should be possible with a bit of javascript-fu.

]]> http://33bits.org/2008/11/12/57/#comment-27 Fred Thu, 27 Nov 2008 04:23:57 +0000 http://33bits.org/2008/11/12/57/#comment-27 Another interesting detail is that Lending Club's "Member map" on their home page seems to place members by ZIP, rather than by city and state (which is on the listings). More fodder for baddies trying to root out users' identities. Another interesting detail is that Lending Club’s “Member map” on their home page seems to place members by ZIP, rather than by city and state (which is on the listings). More fodder for baddies trying to root out users’ identities.

]]>