I am trying to run the Bayes Learner on a directory of spam that ive saved
from outlook express, there are 2000 spam messages in the folder, 1850-1900
are tagged by spam assassin which i believe doesnt cause a problem. outlook
express saves files as ".eml" when it saves them

yet when i ran the learner on the directory the first time it learned from
only 4 messages.

after a lot of messin around i found that if i changed the name from
*****SPAM***** RANDOM SPAM SUBJECT to anything at all eg "therewlhk" it
worked and the learner learned from that message.

i think the problem may be that windows does not allow "*" characters in the
filename so it comes out as "_____SPAM_____" then the learner doesnt pick
this out as a tagged message or something, but that still doesnt aggount for
all the messages.

has abyone else had the same problem and how have they worked round it

help would be appreciated

pete



-------------------------------------------------------
This SF.NET email is sponsored by: eBay
Great deals on office technology -- on eBay now! Click here:
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk

Reply via email to