Spam Folder: Difference between revisions

Back up to the Email Features Category
From AAISP Support Site
m (AA-Andrew moved page Default Folders to Spam Folder: Feature updated to reflect new way of learning spam and ham)
mNo edit summary
Line 1: Line 1:
<indicator name="Faults">[[File:Menu-emailadd.svg|link=:Category:Email_Features|30px|Back up to the Email Features Category]]</indicator>
<indicator name="Faults">[[File:Menu-emailadd.svg|link=:Category:Email_Features|30px|Back up to the Email Features Category]]</indicator>
We create some special folders manually to help with dealing with spam.


When connecting to our email service over IMAP we will create a special folder named spam (lowercase). If you delete the spam folder it will be re-created next time you log in.
*spam
*learnspam
*learnham


The service will then move messages that has been classified as spam in to this folder. This helps
Note that these names are case sensitive.
keep your inbox free of spam and you should keep an eye on this folder in case there are messages
that are not spam.


Our spam checking system performs many checks on a
===spam===
message to decide whether it is spam or not. One of the
Messages which the system has marked as spam will be placed
checks it can do, is to compare the message with previous
in to the spam folder rather than your inbox. -this helps
messages. -this is called Bayesian Learning.
keep your inbox free of spam.


==Teaching the system==
You should keep an eye on this folder in case there are messages
that are not spam. -if you do find such messages, then move then
If our system is marking non-spam messages as spam, and not marking spam messages, then one way to improce this is
to teach the Bayesian Learning system about messages that have been miss-classified. This is done in the following ways:
to the learnham folder which is in your Inbox.


If you delete the spam folder, then spam messages will go to
*Move a message '''OUT''' of the spam folder to tell the system it is '''NOT''' spam
*Move a message '''IN''' to the spam folder to tell the system it '''IS''' spam
your Inbox folder instead.


If you make a mistake, just move the message back.
===learnspam and learnham===

The special spam folders leanspam and
are folders that you can move messages in to so as to help
the Bayesian system deal with your email.

TEACHING A MESSAGE AS SPAM:

If you have a message which is spam, but the system
has not marked as spam, then you can move the message to
your spam folder. This will get learnt as it's moved.

You can also move the message in to the learnspam folder
for it to get learnt overnight, it will then be moved
back to your spam folder once learnt.

TEACHING A MESSAGE AS NOT-SPAM (HAM):

If you have a message that has been marked as spam, but
is not a spam message, then you can do move it out
of the spam folder and in to your Inbox and it will be
learnt as it's moved.



Here is how the look in webmail:

[[File:Email-default-folders.png]]

(The 'Junk; folder is actually called 'spam', but in this case, Roundcube displays the folder as 'Junk' instead of 'spam'.)


The process of learning doesn't guarantee that the system will be correct next time a similar message is received, but it should help.


[[Category:Email Features]]
[[Category:Email Features]]

Revision as of 12:14, 2 July 2019


When connecting to our email service over IMAP we will create a special folder named spam (lowercase). If you delete the spam folder it will be re-created next time you log in.

The service will then move messages that has been classified as spam in to this folder. This helps keep your inbox free of spam and you should keep an eye on this folder in case there are messages that are not spam.

Our spam checking system performs many checks on a message to decide whether it is spam or not. One of the checks it can do, is to compare the message with previous messages. -this is called Bayesian Learning.

Teaching the system

If our system is marking non-spam messages as spam, and not marking spam messages, then one way to improce this is to teach the Bayesian Learning system about messages that have been miss-classified. This is done in the following ways:

  • Move a message OUT of the spam folder to tell the system it is NOT spam
  • Move a message IN to the spam folder to tell the system it IS spam

If you make a mistake, just move the message back.

The process of learning doesn't guarantee that the system will be correct next time a similar message is received, but it should help.