On Tue, 24 Jul 2018, Nick Bright wrote:
On 7/24/2018 9:58 AM, John Hardin wrote:
However, unless you *really* trust the people who are providing training
data, you don't train on the submissions without first reviewing them.
Therefore, forwarding as an RFC-822 attachment isn't a deal-killer.
Nick Bright wrote:
On 7/23/2018 11:49 PM, Bill Cole wrote:
The goal is to get a copy of the message that is identical to what SA
saw when it arrived. For IMAP users, this is easiest to get with a
'missed spam' mailbox into which users can move messages for learning.
If you must rely on forward
On 24 Jul 2018, at 13:39, Nick Bright wrote:
On 7/23/2018 11:49 PM, Bill Cole wrote:
The goal is to get a copy of the message that is identical to what SA
saw when it arrived. For IMAP users, this is easiest to get with a
'missed spam' mailbox into which users can move messages for
learning.
On 7/24/2018 9:58 AM, John Hardin wrote:
However, unless you *really* trust the people who are providing
training data, you don't train on the submissions without first
reviewing them.
Therefore, forwarding as an RFC-822 attachment isn't a deal-killer.
You can review the submission and if you
On 7/24/2018 1:47 AM, Pedro David Marco wrote:
>On Tuesday, July 24, 2018, 1:38:59 AM GMT+2, Nick Bright
wrote:
>So I ask: what is the best practice for learning submissions when
using site-wide bayes?
Nick, do all your users use the same MUA? There are some user level
"plu
On 7/23/2018 11:49 PM, Bill Cole wrote:
The goal is to get a copy of the message that is identical to what SA
saw when it arrived. For IMAP users, this is easiest to get with a
'missed spam' mailbox into which users can move messages for learning.
If you must rely on forwarded submissions, make
Hi,
>> The problem I'm trying to solve is "how to implement a training system on
>> my server".
>
> I'd suggest a manual review step before feeding the messages to Bayes.
>
> You **WILL** get users reporting all kinds of "unwanted today because
> Reasons" but otherwise legitimate email as spam.
W
On Mon, 23 Jul 2018 18:38:48 -0500
Nick Bright wrote:
> When requesting submissions from users for use with sa-learn, if they
> are going to forward the message somewhere; is it best for that to be
> forwarded as an attachment, or forwarded inline?
Don't use forward inline, the choice is betwe
On Mon, 23 Jul 2018, Nick Bright wrote:
On 7/23/2018 8:10 PM, Reindl Harald wrote:
what exactly is the problem right-click on the attachments, save
them to files and drag&drop them to the imap training folder?
What on earth is even right about it? I'm not going to do that for hundreds
or even
On Mon, 23 Jul 2018, Nick Bright wrote:
On 7/23/2018 7:30 PM, Reindl Harald wrote:
provide imap-shared folders and code a script which fetches the
raw-messages and fires sa-learn to the eml-files
*never* train inline-forwardings
And when that isn't an option, for example with POP3 clients?
On Tue, 24 Jul 2018 10:34:42 -0400
Kris Deugau wrote:
> Kris Deugau wrote:
> > Nick Bright wrote:
> >
> >> The problem I'm trying to solve is "how to implement a training
> >> system on my server".
> >
> > I'd suggest a manual review step before feeding the messages to
> > Bayes.
> >
> > Yo
Kris Deugau wrote:
Nick Bright wrote:
The problem I'm trying to solve is "how to implement a training system
on my server".
I'd suggest a manual review step before feeding the messages to Bayes.
You **WILL** get users reporting all kinds of "unwanted today because
Reasons" but otherwise leg
Nick Bright wrote:
The problem I'm trying to solve is "how to implement a training system
on my server".
I'd suggest a manual review step before feeding the messages to Bayes.
You **WILL** get users reporting all kinds of "unwanted today because
Reasons" but otherwise legitimate email as spa
Nick Bright schrieb am 24.07.2018 um 01:38:
So I ask: what is the best practice for learning submissions when
using site-wide bayes?
From what I learnt about best practice:
- before implementing spam-learning based on user-submissions, figure
out how educated your users are with
On Tuesday, July 24, 2018, 6:50:13 AM GMT+2, Bill Cole
wrote:
> Learning ham is harder
Totally agree Bill, unless you use Microsoft technics...: send everything to
spam folder and if moved to inbox by user then... it is ham!
-PedroD
>On Tuesday, July 24, 2018, 1:38:59 AM GMT+2, Nick Bright
wrote:
>So I ask: what is the best practice for learning submissions when using
>site-wide bayes?
Nick, do all your users use the same MUA? There are some user level "plug-ins"
that may be configured to sen
ing.
So I ask: what is the best practice for learning submissions when
using site-wide bayes?
The goal is to get a copy of the message that is identical to what SA
saw when it arrived. For IMAP users, this is easiest to get with a
'missed spam' mailbox into which users can move messa
On Mon, 23 Jul 2018, Nick Bright wrote:
On 7/23/2018 7:55 PM, Reindl Harald wrote:
and even if - whats the point to store the surrounding messages in the
corpus which you should keep forever if you need rebuild from scratch
later?
what is the problem you try to solveand why can't you just stor
On 7/23/2018 8:10 PM, Reindl Harald wrote:
what exactly is the problem right-click on the attachments, save
them to files and drag&drop them to the imap training folder?
What on earth is even right about it? I'm not going to do that for
hundreds or even thousands of submissions sent in by users
On 7/23/2018 7:55 PM, Reindl Harald wrote:
and even if - whats the point to store the surrounding messages in the
corpus which you should keep forever if you need rebuild from scratch
later?
what is the problem you try to solveand why can't you just store the attachment
instead the whole mail
On 7/23/2018 7:49 PM, Reindl Harald wrote:
surely, just right-click on the attachment and save it to a raw-message
(eml-file)
So that's a "no" (sa-learn doesn't know how to 'right click' an attachment).
--
---
- Nick Bright
On 7/23/2018 7:30 PM, Reindl Harald wrote:
provide imap-shared folders and code a script which fetches the
raw-messages and fires sa-learn to the eml-files
*never* train inline-forwardings
And when that isn't an option, for example with POP3 clients?
Is it possible to train on attached forwar
Learning from a mailbox of my own spam (with full headers - the actual
mails) is quite different from users *forwarding* spam for training.
So I ask: what is the best practice for learning submissions when using
site-wide bayes?
--
---
23 matches
Mail list logo
