Feature Requests item #1233040, was opened at 2005-07-05 15:58
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=498106&aid=1233040&group_id=61702

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Outlook
Group: None
Status: Open
Priority: 5
Submitted By: Paul Grunwald (pgrunwald)
Assigned to: Mark Hammond (mhammond)
Summary: Graphic Hash score?

Initial Comment:
I upgraded to the latest version and the majority of the 
spam that is getting through is a graphic that is an ad for 
perscription drugs.  The bottom of the spam is a random 
word list to confuse the lexical scores.

Would it be possible to develop another score that perhaps 
took a bit hash of the graphic in the spam and then looked 
for the same hash on subsequent messages?  A more 
complicated solution would be to look for "similar" pictures 
but I would think this would be an order of magnitude more 
complicated than a simple hash or other calculation.  Other 
things that could contribute to the score could be size, 
shape, color depth, percentage of white space or 
background color, etc.

I know that it would be somewhat easy to put in a random 
bit in the graphic to cause a different hash but it would put 
a CPU tax on the spammers to randomize the graphics for 
every message.  


Just a thought,
Paul

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=498106&aid=1233040&group_id=61702
_______________________________________________
Spambayes-bugs mailing list
[email protected]
http://mail.python.org/mailman/listinfo/spambayes-bugs

Reply via email to