Le 12/01/2011 23:02, Mahmoud Khonji a écrit : > I would highly appreciate if anyone is able to send me his SPAM/Ham email > collection.
sigh. if you can't understand what "privacy" means, then you are part of the problem. > > I need it to train and test classifiers. you need to train with _your_mail. do not train with somebody else's mail. one of the defence args is that attackers can't guess your setup. if every one of us uses the same corpus then it'll be easy for an attacker to get around. > > The issue with available corpus is that they are outdated. They generally > date back in 2005, and lot has changed since then -- We've got SPAMers with > spell checkers at least! > Wondering why did some folks take the initiative in ~2005, and then stopped > contributing to the corpus? I notice the need is still there, and many are > desperate to this kind of corpus. >
