3. Training the Bayesian Filter

The latest version of SpamAssassin has a Bayesian filter that can improve SpamAssassin's ability to distinguish the spam from the ham. SpamAssassin will sometimes mark spam as ham or vice versa. When this happens, you can tell SpamAssassin's Bayesian filter to "relearn" a mail as spam or ham.

On netra, use the sa-learn command to train the Bayesian filter. Here are some examples:

sa-learn --forget --ham --mbox /var/spool/mail/netid

This will tell SpamAssassin that all messages in your inbox are not spam. The --forget flag tells it to forget the message if it has learned it before. Without this flag it will ignore messages that it has already marked as spam or ham.

sa-learn --forget --spam --mbox /home/netid/mail/NewSpam

This tells SpamAssassin that all messages in the NewSpam folder should be relearned as Spam messages. If you moved messages from your inbox to this directory because SpamAssassin didn't catch the message as Spam, this command will train the Bayesian filter accordingly.

From the sa-learn manpage:

NAME
       sa-learn - train SpamAssassin's Bayesian classifier

SYNOPSIS
       sa-learn [options] --file message

       sa-learn [options] --mbox mailbox

       sa-learn [options] --dir directory

       sa-learn [options] --single < message

       Options:

        --ham                             Learn messages as ham (non-spam)
        --spam                            Learn messages as spam
        --forget                          Forget a message
        --rebuild                         Rebuild the database if needed
        --force-expire                    Force an expiry run, rebuild every time
        -f file, --folders=file           Read list of files/directories from file
        --dir                             Learn a directory of RFC 822 files
        --file                            Learn a file in RFC 822 format
        --mbox                            Learn a file in mbox format
        --showdots                        Show progress using dots
        --no-rebuild                      Skip building databases after scan
        -L, --local                       Operate locally, no network accesses
        -C file, --config-file=file       Path to standard configuration dir
        -p prefs, --prefs-file=file       Set user preferences file
        -D, --debug-level                 Print debugging messages
        -V, --version                     Print version
        -h, --help                        Print usage message

See the sa-learn manpage on netra for more information.