how_to:training_the_spam_filter

Training the VPOP3 Spam Filter

The VPOP3 spam filter has a couple of mechanisms which can be used to 'train' it to distinguish between spam and not-spam.

Bayesian Filter

The Bayesian filter this looks at words in messages sent and received and their probabilities of being contained in spam and not-spam, and works out an overall probability of the message being spam or not-spam based on the combination of those probabilities.

  • Every time you send a message, the words in that message are learnt as being words found in not-spam.
  • Every time you receive a message which VPOP3 decides is spam, the words in that message are learnt as being words found in spam.
  • Every time you receive a message which VPOP3 decides is not spam, the words in that message are learnt as being words found in not-spam.
  • If you receive a message which is spam, which the spam filter did not detect, you can forward it to spam@<your domain>. The spam filter will then un-learn the words in that message as being words in not-spam, and re-learn them as being words found in spam. (Note that although forwarding the message will alter the message slightly usually this is not significant).
  • If you receive a message which is spam and view it in Webmail you can mark the message as spam from there and VPOP3 will un-learn the words as being not-spam, and re-learn them as spam.
  • If you receive a message which is not spam, but which the spam filter incorrectly detected as spam, when you release the message from the spam quarantine, the spam filter will un-learn the words as being found in spam, and re-learn them as words being found in not-spam.

Whitelist

The Whitelist contains email addresses which are 'trusted'. Messages received from any of these email addresses will be far more likely to be treated as not-spam than messages received from elsewhere.

  • Every time you send a message, the recipients of that message are added to the whitelist (if they are not already there)
  • Every time you release a message from the spam quarantine, you can choose to add the sender of that message to the whitelist.
  • The administrator can manually add entries to the whitelist

Note that VPOP3 will never automatically add addresses to the 'Blacklist'. This is because spammers often forge email addresses, so automatically adding entries to the Blacklist runs the risk of adding legitimate senders to that list. Also, it rarely is of any benefit, since spammers rarely re-use their forged addresses, so a big Blacklist just slows down the spam filtering for only minor benefit.

how_to/training_the_spam_filter.txt · Last modified: 2018/11/14 10:45 by 127.0.0.1