home posts contact

New paper: Frequency analysis of second-level domain names and detection of pseudo-random domain generation

13 Jun 2010

Its long title I know. While doing some other related personal research, I wrote this excerpt:

We can also look at the randomness of a domain name. For example ajksudmapx.com looks to be random for a human. To reverse randomness would be similarly as difficult as reversing a hashing algorithm. We could look at how domain names are spelt. We could look at the distribution of letters, assuming

You can see that I didn’t finish the sentence. I put what I was working on hold and thought I would investigate the possibility of detecting randomly-generated domain names. It is a bit rough and not completely polished, but I would have never gotten around to putting it online otherwise.

Get the PDF here

comments powered by Disqus

2014 - Ryan Doyle - Powered by Jekyll