<br><br>
<div class="gmail_quote">On Tue, Mar 25, 2008 at 9:31 PM, Agutin Gianni <<a href="mailto:agustingianni@gmail.com">agustingianni@gmail.com</a>> wrote:<br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div class="Ih2E3d">-----BEGIN PGP SIGNED MESSAGE-----<br>Hash: SHA1<br><br></div>I think we have already discussed this topic, and someone said we could<br>use pictures of cats and other animals and ask the user to count the<br>
number of cats on the photos.<br><br>Microsoft is working on this, it looks promising.<br><br><a href="http://research.microsoft.com/asirra/" target="_blank">http://research.microsoft.com/asirra/</a><br>
<div>
<div></div></div></blockquote>
<div> </div></div>
<div class="gmail_quote">As cool as ASIRRA is, and as awesome as it is that they help find homes for pets, it is more or less "a better CAPTCHA." I took the original post as a request for a manual for bot-detection techniques in addition to CAPTCHAs. I don't know of anything in one place on this topic, although I can think of things like Bayesian filters for the spam application as maybe a place to start.</div>
<div class="gmail_quote"> </div>
<div class="gmail_quote">Incidentally, this paper just showed up on <a href="http://eprint.iacr.org">eprint.iacr.org</a> . The author claims an automatic classifier between cats and dogs that can pass a 12-image ASIRRA challenge 10.3% of the time: </div>
<div class="gmail_quote"> </div>
<div class="gmail_quote">Machine Learning Attacks Against the ASIRRA CAPTCHA</div>
<div class="gmail_quote">Philippe Golle</div>
<div class="gmail_quote"><a href="http://eprint.iacr.org/2008/126">http://eprint.iacr.org/2008/126</a></div>
<div class="gmail_quote"> </div>
<div class="gmail_quote">-David Molnar</div>