Re: train-good and train-spam; cron error
Anthony Campbell <ac-MeA/[email protected]>
| Newsgroups | gmane.mail.spam.spamprobe.general |
|---|---|
| Message-ID | <[email protected]> |
On 10 Dec 2004, Brian Burton wrote: > John Chandler wrote: > > The documentation still recommends using "good" and "spam" > > for the initial training and for later corrrective training. > > If you use train for incoming email then you should use train-good and > train-spam for corrections. Using the good/spam commands would not hurt > anything but could increase database size. > > > > Would the use of "train-good" or "train-spam" > > in initially training spamprobe > > or in later corrective sessions be appropriate, > > in order better to control the size of the database? > > Once 1.1 is finished (needs documentation changes prior to release) the > proper way to build an initial database will be the auto-train command. > Until then I recomment using the good and spam commands. > > auto-train will do its best to minimize the database size during > training and will train on your saved hams and spams in a balanced way > that should improve starting accuracy. > > [snip] auto-train seems to work pretty well. A few weeks ago my database seemed to go wrong in some unspecified way so I decided to start afresh. I started with auto-train and now, after about 3 weeks, I seem to be back to the original standard of accuracy and, as a bonus, my database is 18M instead of 65M. Admittedly I had a good source of spam to train on, because one of the mailservers I use has a spam filtering system (not nearly as efficient as SP however) so for several days I downloaded all the spam and used that. Anthony -- ac-MeA/[email protected] || http://www.acampbell.org.uk using Linux GNU/Debian || for book reviews, electronic Windows-free zone || books and skeptical articles ------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/