On Tue, 2006-03-07 at 02:34 -0800, Loren Wilton wrote: > X-Spam-Status: No, score=4.863 tagged_above=-100 required=6.3 > tests=[BAYES_50=0.001, HTML_IMAGE_ONLY_12=1.867, HTML_MESSAGE=0.001, > HTML_SHORT_LINK_IMG_1=0.948, RCVD_IN_SORBS_DUL=2.046] > > > How can I increase the score of this message? > > 1. Train Bayes. It doesn't know if this is ham or spam. You can get > another 3 points or so (depending on version) if Bayes was convinced this > was spam. > Thank you for the reply. Maybe I should train the bayes as you said. Will have to see how to correctly train it, but I'm afraid it get poisoned since this is a site wide setup, not on user bases!
> 2. It looks like you may be on 2.63 or so. Move to 3.0.5 or 3.1.x, and > make sure the URIBL rules are enabled, as they are by default. > I'm running spamassassin 3.1.0. Do you mean URIBL as network tests? I have pyzor, DCC and razor tests enabled. > 3. With only the spam score and nothing else from the message, it is > impossible to make any other useful guesses. > :( > Loren > Sincerely, Yousef Raffah Senior Systems Administrator SSIS - The Savola Group -- Aren't you using Firefox? Get it at getfirefox.com yousef.raffah.com
signature.asc
Description: This is a digitally signed message part
