rewrite_header Subject and Bayes

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

rewrite_header Subject and Bayes

Palvelin Postmaster
Silly question or not, here goes:

I prepend my spam emails’ subject fields with a specific string to indicate spam, like many do, I presume. Will that string get noticed by bayes and if so, should I do something to prevent it?

--
Palvelin.fi Hostmaster
[hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: rewrite_header Subject and Bayes

Matus UHLAR - fantomas
On 30.05.18 15:12, Palvelin Postmaster wrote:
>I prepend my spam emails’ subject fields with a specific string to indicate
> spam, like many do, I presume.  Will that string get noticed by bayes and
> if so, should I do something to prevent it?

most probably, yes.

However, not by your bayes, unless you check for spamminess, tag and check
again...


--
Matus UHLAR - fantomas, [hidden email] ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
Save the whales. Collect the whole set.
Reply | Threaded
Open this post in threaded view
|

Re: rewrite_header Subject and Bayes

Palvelin Postmaster


> On 30 May 2018, at 15:21, Matus UHLAR - fantomas <[hidden email]> wrote:
>
> On 30.05.18 15:12, Palvelin Postmaster wrote:
>> I prepend my spam emails’ subject fields with a specific string to indicate
>> spam, like many do, I presume.  Will that string get noticed by bayes and
>> if so, should I do something to prevent it?
>
> most probably, yes.
>
> However, not by your bayes, unless you check for spamminess, tag and check
> again...

Yes, forgot to mention I store tagged spam messages and run sa-learn on them to teach spam/ham.
Reply | Threaded
Open this post in threaded view
|

Re: rewrite_header Subject and Bayes

Matus UHLAR - fantomas
>> On 30.05.18 15:12, Palvelin Postmaster wrote:
>>> I prepend my spam emails’ subject fields with a specific string to indicate
>>> spam, like many do, I presume.  Will that string get noticed by bayes and
>>> if so, should I do something to prevent it?

>> On 30 May 2018, at 15:21, Matus UHLAR - fantomas <[hidden email]> wrote:
>> most probably, yes.
>>
>> However, not by your bayes, unless you check for spamminess, tag and check
>> again...

On 30.05.18 15:41, Palvelin Postmaster wrote:
>Yes, forgot to mention I store tagged spam messages and run sa-learn on them to teach spam/ham.

it's better to keep spam sign in X-Spam-* headers, which are ignored by
spamassassin.
--
Matus UHLAR - fantomas, [hidden email] ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
On the other hand, you have different fingers.