Bayesian has caught all of these so far for me. My only concern is that my corpus may get poisoned somehow... is this feasible or will it be ok?
Hill, Brett wrote: > Anyone else seeing this new spam variant? Any ideas how to fight it. > Far as I can tell, RBL checks are catching most of them, but some manage > to get through. They look to be 4 lines long and they're in html > format. Also, none of them seem to be alike. Seems like they all start > out with a website address and then some sort of message after it. > > I tried to send an edited example to the list, but sourceforge's email > server blocked it citing spam. I'll try again. > > The email would look something like this: > > http://website.com > Hi username > Buy such and such to make yourself large > A made-up person's name > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > > > ------------------------------------------------------------------------ > > _______________________________________________ > Assp-user mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/assp-user ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Assp-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/assp-user
