[liberationtech] On the Feasibility of Internet-Scale Author Identification

[Steve Weis]

There is some interesting research coming out from a team of Stanford &
Berkeley researchers about large-scale de-anonymization of blog posts based
on writing style, i.e. stylometry.

For a given blog post, the researchers were able to positively identify an
individual author from among 100,000 possibilities 20% of the
time. However, their method does not work if authors deliberately obfuscate
their writing style.

Here's the paper draft:
http://randomwalker.info/publications/author-identification-draft.pdf

And a blog post about it:
http://33bits.org/2012/02/20/is-writing-style-sufficient-to-deanonymize-material-posted-online/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/liberationtech/attachments/20120220/a2efad99/attachment.html>