EmailDiscussions.com  

Go Back   EmailDiscussions.com > Email Service Provider-specific Forums > FastMail Forum
Register FAQ Members List Calendar Search Today's Posts Mark Forums Read
Stay in touch wirelessly

FastMail Forum All posts relating to FastMail.FM should go here: suggestions, comments, requests for help, complaints, technical issues etc.

Reply
 
Thread Tools
Old 6 Sep 2017, 10:07 PM   #1
rabarberski
Master of the @
 
Join Date: Nov 2006
Location: Ghent, Belgium
Posts: 1,027
Strategy for old mails in SPAM (auto-learning) folder?

I currently have about 850 messages in my fastmail spam folder.
There were times I kept all of them (> 5k ?), but because of inbox size issues I had to clean it about a year ago.

Again, I would like that my spam folder doesn't grow to large (by deleting messages), but still is effective in spam filtering (using the fastmail auto-learning feature).

Is there a particularly good strategy to keep the size of my spam folder under control?
Just remove the ones that are over a year old?
Or maybe the learning is even more effective if I just keep the ones that are e.g. maximum 6 months old, so more recent spam message relatively weight more in the whole personal spam "database"?
rabarberski is offline   Reply With Quote

Old 6 Sep 2017, 11:20 PM   #2
BritTim
The "e" in e-mail
 
Join Date: May 2003
Location: mostly in Thailand
Posts: 3,084
Keeping the learned spam messages under a year old is actually an excellent approach. It should be quite sufficient to tune your personal Bayes database (should it ever need to be reset) and using the advanced preferences for a folder, you can set an auto purge limit of 365 days. Note that (although I doubt it will be that important) if you do decide a particular message is important to keep, you can pin the message to prevent auto purge from discarding it.
BritTim is offline   Reply With Quote
Old 7 Sep 2017, 12:11 AM   #3
rabarberski
Master of the @
 
Join Date: Nov 2006
Location: Ghent, Belgium
Posts: 1,027
thx, good tip.
i've enable the 365 days purging
rabarberski is offline   Reply With Quote
Old 7 Sep 2017, 03:23 PM   #4
n5bb
Intergalactic Postmaster
 
Join Date: May 2004
Location: Irving, Texas
Posts: 8,917
Arrow How the spam filter really works

Leaving messages in the Spam folder does not affect spam filtering! I think you misunderstand how the Fastmail spam filtering system and personal Bayes database work. Refer to this section of the Fastmail help:
https://www.fastmail.com/help/receive/stopspam.html

Until you get >200 spam and >200 non-spam messages reported, your personal Bayes database does not operate. Messages are reported to the Bayes database as follows:
  • Messages which are permanently deleted from the Spam folder are marked as spam in the database when they are permanently deleted. You can do this by manually permanently deleting messages in the Spam folder or using the auto-purge property for that folder.
  • Messages which are missed by the spam filter but marked by the user with the Report Spam button are automatically moved to the Spam folder and immediately marked as spam in the database.
  • Messages which are accidentally moved by the spam filter to the Spam folder but are then marked as Not Spam by the user are automatically moved to the Inbox folder and immediately marked as non-spam in the database.
  • Messages in folders whose property is set for spam learning are treated as follows. One time per day (I think it's usually in the night) all of your folders are scanned. If the folder is set for spam learning, any message which was added to that folder since the last daily scan is learned as specified in the folder properties. The message is only learned one time (the first time it is scanned after the message is added to that folder).
  • Fastmail strongly suggests that the Spam folder not be set up for automatic learning to prevent spam learning loops (read the Help for more details).
So if you have a pile of message in your Spam folder which were moved there by the automatic spam filter (or a user rule), those messages will not affect the spam filter until they are permanently deleted (unless you ignore the Fastmail directions to not set the Spam folder to be learned). The user flow which is best is:
  • Look at your Spam folder when it's convenient. Mark any non-spam as Not Spam and Delete Permanently any spam. Notice that when you in the Spam folder, the Delete button is changed to Delete Permanently.
  • If you rarely look at the spam folder you might miss good messages improperly filed into the spam folder. So it's good to check the Spam folder before any auto-purge folder setting permanently deletes the message. But remember that if you set the auto-purge to 365 days, the messages automatically filed into the Spam folder will not affect your personal Bayes spam filter until they have been there for one year. I don't thing that's what you intended!
Bill
n5bb is offline   Reply With Quote
Old 7 Sep 2017, 07:44 PM   #5
rabarberski
Master of the @
 
Join Date: Nov 2006
Location: Ghent, Belgium
Posts: 1,027
Wow, thanks Bill.
Ind-dept explanation.
The "user flow which is best" was very useful. I am adapting my flow already
rabarberski is offline   Reply With Quote
Old 8 Sep 2017, 09:43 AM   #6
n5bb
Intergalactic Postmaster
 
Join Date: May 2004
Location: Irving, Texas
Posts: 8,917
Yeah ... I was confused about this until I re-read the latest help. I now understand the trap Fastmail is worried we might get into if we mark the Spam folder as learn-as-spam. So they now don't learn messages as spam unless you do one of these things:
  • Mark an undesired message in your Inbox (or some other normal folder) which evaded the spam filer as spam. I believe this immediately marks it as spam in the database.
  • Permanently delete messages from the Spam folder. Fastmail is assuming that we will look at the Spam folder to verify there aren't any ham (non-spam) in there before we delete the messages. I believe that the auto-purge folder property will do this also.
    • All of the permanently deleted messages from the Spam folder are learned as spam. So it's very important to not delete any non-spam messages which are accidentally in that folder! You need to mark ham (in the Spam folder) as Not Spam so they are learned correctly.
Bill
n5bb is offline   Reply With Quote
Old 8 Sep 2017, 04:33 PM   #7
Terry
The "e" in e-mail
 
Join Date: Jul 2002
Location: VK4
Posts: 2,995
Bill how do we reset the filter (Spam count) there is no delete or Zero button.
Terry is offline   Reply With Quote
Old 9 Sep 2017, 11:47 AM   #8
n5bb
Intergalactic Postmaster
 
Join Date: May 2004
Location: Irving, Texas
Posts: 8,917
Quote:
Originally Posted by Terry View Post
Bill how do we reset the filter (Spam count) there is no delete or Zero button.
You can't. But you can train the Bayes database correctly in the future. If you believe you have improperly trained your database, you can file a support request and I'm pretty sure Fastmail staff can reset it.

Bill
n5bb is offline   Reply With Quote
Old 9 Sep 2017, 01:49 PM   #9
Terry
The "e" in e-mail
 
Join Date: Jul 2002
Location: VK4
Posts: 2,995
Quote:
Originally Posted by n5bb View Post
You can't. But you can train the Bayes database correctly in the future. If you believe you have improperly trained your database, you can file a support request and I'm pretty sure Fastmail staff can reset it.

Bill
Yes that is what I thought as I had trouble trying to find the button.... I just thought F/m had hid it up somewhere else.
Terry is offline   Reply With Quote
Old 12 Sep 2017, 11:59 PM   #10
jhollington
Essential Contributor
 
Join Date: Apr 2008
Posts: 371
Quote:
Originally Posted by n5bb View Post
Yeah ... I was confused about this until I re-read the latest help. I now understand the trap Fastmail is worried we might get into if we mark the Spam folder as learn-as-spam.
This is kind of an interesting point, and it took a moment or two to wrap my head around where they were going with this, but I can see how it makes sense.

However, I've had my Spam folder set as learning spam for years, and haven't really had any problems with it. Part of the reason for me is that I primarily use Apple Mail so it's just an easier way to deal with getting stuff taught as spam — I very rarely use the web interface. I also strongly suspect that FastMail won't recognize deletion of mails from the spam folder via IMAP —*probably also not surprising since most IMAP clients will do a move-to-trash rather than an expunge. I realize I could point Apple Mail to use any arbitrary folder as its own "Junk Mail" folder, but again, I haven't really had a problem and I don't relish the idea of having two spam folders

In my case, however, I've also set several other folders to learn "not-spam," including my Inbox and Archive folders. So in the event that something does get into a false positive loop, I could very well be misunderstanding how the Bayes training works, but I've always assumed the process of moving it back to the Inbox or Archive folders would effectively reverse the spam training for that message the next time those folders are scanned.

Further, since I check my Inbox every single day and the scanning only occurs on a cycle, there doesn't seem to be a high risk of false negatives (e.g. learning spam messages as legit) from having that set as a not-spam training folder — chances are anything that lands in there that's spam will be dealt with before the next FastMail scan of that folder.

Now, all of that having been said, I should note that I've been at this for a very long time with FastMail, so my Bayes database numbers for both spam and non-spam are into the tens of thousands. It might be a very different scenario for somebody who is just getting started.
jhollington is offline   Reply With Quote
Old 13 Sep 2017, 12:36 AM   #11
jhollington
Essential Contributor
 
Join Date: Apr 2008
Posts: 371
Here's something else from FastMail's Improving Spam Protection help article that I didn't realize (emphasis mine):

Quote:
We also automatically train this with spam you've deleted permanently from your spam folder, and non-spam you've moved to your Archive folder or replied to.
So there's no need to set the Archive folder as "non-spam learning" as it looks like this is done automatically. Also makes logical sense that if you're replying to a message, chances are it's not spam.

Even though I haven't really been having any serious problems with my spam filtering, all of this is making me rethink my approach.... particularly useful is the knowledge that the spam training database gets updated when a message is expunged from the spam folder, even with auto-purge.
jhollington is offline   Reply With Quote
Old 13 Sep 2017, 08:39 AM   #12
neilj
Cornerstone of the Community
 
Join Date: Apr 2004
Location: Melbourne
Posts: 971

Representative of:
Fastmail.fm
Quote:
We also automatically train this with spam you've deleted permanently from your spam folder, and non-spam you've moved to your Archive folder or replied to.
This only applies to the web interface/our apps. If you're just using IMAP, the only way to train the spam filter is by setting the auto-learning on folders.

Neil.
neilj is offline   Reply With Quote
Old 13 Sep 2017, 08:44 AM   #13
jhollington
Essential Contributor
 
Join Date: Apr 2008
Posts: 371
Quote:
Originally Posted by neilj View Post
This only applies to the web interface/our apps. If you're just using IMAP, the only way to train the spam filter is by setting the auto-learning on folders.
Ah, okay, I get it. Thanks for the clarification.
jhollington is offline   Reply With Quote
Reply


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is Off
HTML code is Off
Forum Jump


All times are GMT +9. The time now is 12:32 AM.

 

Copyright EmailDiscussions.com 1998-2022. All Rights Reserved. Privacy Policy