Spam From Idiots

I have a mailbox reserved for spam, where the headers are
stored for possible future study.

wahoo77@yahoo.com

Been noticing a lot of subjects like this:

Subject: aardvark test oceans mega...
Subject: So she walked to the corn...
Subject: slimy athabasca never lov...
Subject: On This Fine Winter Day S...
Subject: generous outage if and th...
Subject: yellow first Commonwealth...

If I ever saw those subjects, or anything like them, in my
real mailboxes, I'd dump them without a second thought.

And I find it hard to believe that _anyone_ would even bother
opening them.

I think the spammers are like the idiot trolls: They just aren't
paying attention and don't notice that no one is reading their
crap. Especially their targets, whom they imagine that they
are devastating, I'm sure.

They spam and spam and spam and never make a cent, and can't
figure out why.

Clue: If a mail looks like it was sent by an inmate at a
mental hospital, 99.9% of the people that receive it are
going to dump it without opening it.

And the other .1% doesn't have enough money to buy anything.

[Note: I don't read the articles of "Sam" or his numerous
sockpuppets or his 'friends', nor any responses to them, and
haven't for years. He follows me all over the Usenet, and I
still don't read his articles. This _really_ pisses him off.
.]

Alan

--
http://home.earthlink.net/~alanconnor/contact.html
http://home.earthlink.net/~alanconnor/elrav1/cr.html
Other URLs of possible interest in my headers.

Re: Spam From Idiots

This is a MIME GnuPG-signed message. If you see this text, it means that
your E-mail or Usenet software does not support MIME signed messages.
The Internet standard for MIME PGP messages, RFC 2015, was published in 1996.
To open this message correctly you will need to install E-mail or Usenet
software that supports modern Internet standards.

--=_mimegpg-commodore.email-scan.com-28607-1147211471-0005
Content-Type: text/plain; format=flowed; charset="US-ASCII"
Content-Disposition: inline
Content-Transfer-Encoding: 7bit

Usenet Beavis writes:

> I have a mailbox reserved for spam, where the headers are
> stored for possible future study.

Nobody cares, Beavis.

> I think the spammers are like the idiot trolls: They just aren't
> paying attention and don't notice that no one is reading their
> crap.

Right, Beavis. _Now_ you're beginning to understand your predicament.

> [Note: it's not my fault that I'm a complete dumbass. I was dropped on my
> head as a child. See http://www.pearlgates.net/nanae/kooks/ac/ for
> more information]
>
> Beavis



--=_mimegpg-commodore.email-scan.com-28607-1147211471-0005
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQBEYQ7Px9p3GYHlUOIRArwIAJ90qnhqUR6f7gR1E4m0U7WGzxgeIgCf bzTd
zsmD0CykD9+Ys151hG0ZtCQ=
=KDUB
-----END PGP SIGNATURE-----

--=_mimegpg-commodore.email-scan.com-28607-1147211471-0005--

Bayesian training [Was: Spam From Idiots]

* Alan Connor wrote:
> I have a mailbox reserved for spam, where the headers are
> stored for possible future study.
>
> wahoo77@yahoo.com
>
> Been noticing a lot of subjects like this:
>
> Subject: aardvark test oceans mega...
> Subject: So she walked to the corn...
> Subject: slimy athabasca never lov...
> Subject: On This Fine Winter Day S...
> Subject: generous outage if and th...
> Subject: yellow first Commonwealth...
>
> If I ever saw those subjects, or anything like them, in my
> real mailboxes, I'd dump them without a second thought.
>
> And I find it hard to believe that _anyone_ would even bother
> opening them.

Me too. But you just got me thinking about my SpamAssassin bayesian
training.

> I think the spammers are like the idiot trolls: They just aren't
> paying attention and don't notice that no one is reading their
> crap. Especially their targets, whom they imagine that they
> are devastating, I'm sure.

I think the above type of emails, at least the ones I have been getting
where the body is just a bunch of random phrases, are meant to throw off
learning/training type spam filters like bayesian ones.

I'm pretty sure you don't use SA, and I know you use a C/R system, but
please indulge me.

Those emails themselves are not spam (at least by my definition since
they are not selling anything, although they are certainly unsolicited
bulk emails).

So my question is this - should I pass those type of emails on to SA's
bayesian training system, or just delete them?

[snip]



--
Troy Piggins
,-o Ubuntu v5.10 (Breezy Badger): kernel 2.6.12-9-386,
o ) postfix 2.2.4, procmail 3.22, mutt 1.5.11i,
`-o slrn 0.9.8.1/rt (score_color patch), vim 7.0

Re: Bayesian training [Was: Spam From Idiots]

This is a MIME GnuPG-signed message. If you see this text, it means that
your E-mail or Usenet software does not support MIME signed messages.
The Internet standard for MIME PGP messages, RFC 2015, was published in 1996.
To open this message correctly you will need to install E-mail or Usenet
software that supports modern Internet standards.

--=_mimegpg-commodore.email-scan.com-30376-1147223031-0002
Content-Type: text/plain; format=flowed; charset="US-ASCII"
Content-Disposition: inline
Content-Transfer-Encoding: 7bit

Troy Piggins writes:

>
> I'm pretty sure you don't use SA, and I know you use a C/R system, but
> please indulge me.

No, he doesn't. Beavis makes a lot of noise about C/R, but in
http://tinyurl.com/rvvsd he lets the poor, suffering cat out of the bag:

Beavis> Note that the filter I use for Usenet mail (see below) is
Beavis> not a C-R system.

That's the best testament as to the effectiveness of a C/R system.

> So my question is this - should I pass those type of emails on to SA's
> bayesian training system, or just delete them?

Beavis is the last person you want to ask for any technical advice. Here's
a brief capsule summary of Beavis's previous technical tips:

http://tinyurl.com/rvvsd - how to run a spider that masquerades as Intenret
Explorer

http://tinyurl.com/5qqb6 - how to securely log in over the Internet

http://tinyurl.com/ys6z4 - Beavis begs someone to help him "hack my C:\
drive"

http://tinyurl.com/ifrt - Beavis, the virus and trojan expert

See the Beavis FAQ that's regularly posted to this newsfroup.



--=_mimegpg-commodore.email-scan.com-30376-1147223031-0002
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQBEYTv3x9p3GYHlUOIRAvqzAJ9cWRi7rec2Zg9GRbQFOOGvLuMAiwCd GShv
+/O60oscbiuf2bVAQ89moc8=
=+ylo
-----END PGP SIGNATURE-----

--=_mimegpg-commodore.email-scan.com-30376-1147223031-0002--

Re: Bayesian training [Was: Spam From Idiots]

* Sam wrote:
> Troy Piggins writes:
>
[snip]
>> So my question is this - should I pass those type of emails on to SA's
>> bayesian training system, or just delete them?
>
> AC is the last person you want to ask for any technical advice.
[snip]

I wasn't asking him directly. I would've emailed him for that.
I posted on USENET which is a public forum. Yes I know some of the
phrases I posted referred to "his" system etc, but anyone can
read and followup here as you have done.

I was hoping for someone with some knowledge about bayesian systems to
reply. That's why I changed the subject heading. In hindsight maybe I
should've started a new thread but I got lazy.

Do you know anything about bayesian training in relation to my question?

> See the AC FAQ that's regularly posted to this newsfroup.

I haven't seen those deliberately provocative and antagonistic posts of
yours for some time since they have nothing to do with this group.

--
Troy Piggins
,-o Ubuntu v5.10 (Breezy Badger): kernel 2.6.12-9-386,
o ) postfix 2.2.4, procmail 3.22, mutt 1.5.11i,
`-o slrn 0.9.8.1/rt (score_color patch), vim 7.0

Re: Bayesian training [Was: Spam From Idiots]

Troy Piggins writes:

> > Subject: aardvark test oceans mega...
> > Subject: So she walked to the corn...
> > Subject: slimy athabasca never lov...
> > Subject: On This Fine Winter Day S...
> > Subject: generous outage if and th...
> > Subject: yellow first Commonwealth...

> Me too. But you just got me thinking about my SpamAssassin bayesian
> training.
>
> > I think the spammers are like the idiot trolls: They just aren't
> > paying attention and don't notice that no one is reading their
> > crap. Especially their targets, whom they imagine that they
> > are devastating, I'm sure.
>
> I think the above type of emails, at least the ones I have been getting
> where the body is just a bunch of random phrases, are meant to throw off
> learning/training type spam filters like bayesian ones.
>
> I'm pretty sure you don't use SA, and I know you use a C/R system, but
> please indulge me.
>
> Those emails themselves are not spam (at least by my definition since
> they are not selling anything, although they are certainly unsolicited
> bulk emails).
>
> So my question is this - should I pass those type of emails on to SA's
> bayesian training system, or just delete them?

If SpamAssasin is automatically learned them as non-spam ('ham'), then
at least you want learn them to spamassasin as spam.

Re: Bayesian training [Was: Spam From Idiots]

* Kari Hurtta wrote:
> Troy Piggins writes:
>
>> > Subject: aardvark test oceans mega...
>> > Subject: So she walked to the corn...
>> > Subject: slimy athabasca never lov...
>> > Subject: On This Fine Winter Day S...
>> > Subject: generous outage if and th...
>> > Subject: yellow first Commonwealth...
>
>> Me too. But you just got me thinking about my SpamAssassin bayesian
>> training.
>>
>> > I think the spammers are like the idiot trolls: They just aren't
>> > paying attention and don't notice that no one is reading their
>> > crap. Especially their targets, whom they imagine that they
>> > are devastating, I'm sure.
>>
>> I think the above type of emails, at least the ones I have been getting
>> where the body is just a bunch of random phrases, are meant to throw off
>> learning/training type spam filters like bayesian ones.
>>
>> I'm pretty sure you don't use SA, and I know you use a C/R system, but
>> please indulge me.
>>
>> Those emails themselves are not spam (at least by my definition since
>> they are not selling anything, although they are certainly unsolicited
>> bulk emails).
>>
>> So my question is this - should I pass those type of emails on to SA's
>> bayesian training system, or just delete them?
>
> If SpamAssasin is automatically learned them as non-spam ('ham'), then
> at least you want learn them to spamassasin as spam.

They are not learned as ham. I have passed them to sa-learn as spam to
date, but I am not sure I should be doing this. They are not really
spam, they are just random phrases/words set to throw off spam trainers.

Sample body:
-----8<-----
Tomorrow begins the last week of Autumn, said Thorin one day.
And winter comes after autumn, said Bifur.
And next year after that, said Dwalin, and our beards will grow
till they hang down the cliff to the valley before anything happens
here. What is our burglar doing for us?
Since he has got an invisible ring, and ought to be a specially
excellent performer now, I am beginning to think he might go through the
Front Gate and spy things out a bit!
Bilbo heard this-the dwarves were on the rocks just : above the
enclosure where he was sitting-and Good Gracious! he thought, so that
is what they are beginning to think, is it? It is always poor me that
has to get them out : of their difficulties, at least since the wizard
left. Whatever am I going to do? I might have known that something
-----8<-----

--
Troy Piggins RLU#415538
My dogma just ran over my karma

Re: Bayesian training [Was: Spam From Idiots]

Troy Piggins writes:

> * Kari Hurtta wrote:
> > Troy Piggins writes:
> >
> >> > Subject: aardvark test oceans mega...
> >> > Subject: So she walked to the corn...
> >> > Subject: slimy athabasca never lov...
> >> > Subject: On This Fine Winter Day S...
> >> > Subject: generous outage if and th...
> >> > Subject: yellow first Commonwealth...
> >
> >> Me too. But you just got me thinking about my SpamAssassin bayesian
> >> training.
> >>
> >> > I think the spammers are like the idiot trolls: They just aren't
> >> > paying attention and don't notice that no one is reading their
> >> > crap. Especially their targets, whom they imagine that they
> >> > are devastating, I'm sure.
> >>
> >> I think the above type of emails, at least the ones I have been getting
> >> where the body is just a bunch of random phrases, are meant to throw off
> >> learning/training type spam filters like bayesian ones.
> >>
> >> I'm pretty sure you don't use SA, and I know you use a C/R system, but
> >> please indulge me.
> >>
> >> Those emails themselves are not spam (at least by my definition since
> >> they are not selling anything, although they are certainly unsolicited
> >> bulk emails).
> >>
> >> So my question is this - should I pass those type of emails on to SA's
> >> bayesian training system, or just delete them?
> >
> > If SpamAssasin is automatically learned them as non-spam ('ham'), then
> > at least you want learn them to spamassasin as spam.
>
> They are not learned as ham. I have passed them to sa-learn as spam to
> date, but I am not sure I should be doing this. They are not really
> spam, they are just random phrases/words set to throw off spam trainers.

SpamAssasin learns also from mail headers, so it is still possiblity that
SpamAsasin learns something usefull.

/ Kari Hurtta

Re: Bayesian training [Was: Spam From Idiots]

Troy Piggins wrote:
> * Kari Hurtta wrote:
>> Troy Piggins writes:
>>
>>>[snip
>>> So my question is this - should I pass those type of emails on to SA's
>>> bayesian training system, or just delete them?
>> If SpamAssasin is automatically learned them as non-spam ('ham'), then
>> at least you want learn them to spamassasin as spam.
>
> They are not learned as ham. I have passed them to sa-learn as spam to
> date, but I am not sure I should be doing this. They are not really
> spam, they are just random phrases/words set to throw off spam trainers.

That has never stopped me from using those to train a bayesian filter.

It can bloat the bayesian database eventually if you have it learn every
single spam or ham message but that's about it. Feed enough of them to
the filter so that SA correctly identifies them as spam, after that have
it learn more only as needed to keep the bayesian filter updated.

Bayesian systems look at headers separately from the body of the
messages. So there are things in the headers that a bayesian filter can
use that are independent of the contents of the message body.

I use bogofilter myself which can be gotten from bogofilter.sourceforge.net

There is a pretty good FAQ there on how bayesian filters work.

> [snip]

--
Garen

Re: Bayesian training [Was: Spam From Idiots]

* Garen Erdoisa wrote:
> Troy Piggins wrote:
>> * Kari Hurtta wrote:
>>> Troy Piggins writes:
>>>
>>>>[snip
>>>> So my question is this - should I pass those type of emails on to SA's
>>>> bayesian training system, or just delete them?
>>> If SpamAssasin is automatically learned them as non-spam ('ham'), then
>>> at least you want learn them to spamassasin as spam.
>>
>> They are not learned as ham. I have passed them to sa-learn as spam to
>> date, but I am not sure I should be doing this. They are not really
>> spam, they are just random phrases/words set to throw off spam trainers.
>
> That has never stopped me from using those to train a bayesian filter.
>
> It can bloat the bayesian database eventually if you have it learn every
> single spam or ham message but that's about it. Feed enough of them to
> the filter so that SA correctly identifies them as spam, after that have
> it learn more only as needed to keep the bayesian filter updated.
>
> Bayesian systems look at headers separately from the body of the
> messages. So there are things in the headers that a bayesian filter can
> use that are independent of the contents of the message body.
>
> I use bogofilter myself which can be gotten from bogofilter.sourceforge.net
>
> There is a pretty good FAQ there on how bayesian filters work.

Ok, thanks Garen and Kari. Will do some reading/research.

--
Troy Piggins RLU#415538
My dogma just ran over my karma

Re: Bayesian training [Was: Spam From Idiots]

On comp.mail.misc, in <20060510093552@usenet.piggo.com>, "Troy
Piggins" wrote:

Hey Troy!

> * Alan Connor wrote:
>
>> I have a mailbox reserved for spam, where the headers are
>> stored for possible future study.
>>
>> wahoo77@yahoo.com
>>
>> Been noticing a lot of subjects like this:
>>
>> Subject: aardvark test oceans mega... >> Subject: menu> So she walked to the corn... slimy athabasca
>> Subject: never lov... On This Fine Winter Day S... generous
>> Subject: outage if and th... yellow first Commonwealth...
>>
>> If I ever saw those subjects, or anything like them, in my
>> real mailboxes, I'd dump them without a second thought.
>>
>> And I find it hard to believe that _anyone_ would even bother
>> opening them.
>
> Me too. But you just got me thinking about my SpamAssassin
> bayesian training.
>
>> I think the spammers are like the idiot trolls: They just
>> aren't paying attention and don't notice that no one is
>> reading their crap. Especially their targets, whom they
>> imagine that they are devastating, I'm sure.
>
> I think the above type of emails, at least the ones I have been
> getting where the body is just a bunch of random phrases, are
> meant to throw off learning/training type spam filters like
> bayesian ones.

Got it. Thanks.

>
> I'm pretty sure you don't use SA.

Correct.

> and I know you use a C/R system, but please indulge me.

Sure.

>
> Those emails themselves are not spam (at least by my definition
> since they are not selling anything, although they are
> certainly unsolicited bulk emails).
>
> So my question is this - should I pass those type of emails on
> to SA's bayesian training system, or just delete them?

Couldn't tell you, Troy, but someone will.

My hunch would be to delete them. If they are sent to throw off
your filter, why give them a chance?

>
> [snip]
>
>
>

No prob.

[Note: I don't read the articles of "Sam" or his numerous
sockpuppets or his 'friends', nor any responses to them, and
haven't for years. He follows me all over the Usenet, and I
still don't read his articles. This _really_ pisses him off.
.]

Alan

--
http://home.earthlink.net/~alanconnor/contact.html
Other URLs of possible interest in my headers.

Re: Bayesian training [Was: Spam From Idiots]

Postscript:

On comp.mail.misc, in , "Alan Connor" wrote:

>>
>> I'm pretty sure you don't use SA.
>
> Correct.
>
>> and I know you use a C/R system, but please indulge me.
>
> Sure.
>

I should note that a lot of C/R users _do_ use SA as the
conventional spam filter stage in their Challenge-Response
system:

http://home.earthlink.net/~alanconnor/elrav1/cr.html

The 'graymail' gets the challenge.

It's very easy, as you well know, to use procmail to pipe
non-passlisted mail through SA and return it to procmail, scored
by SA, for the C/R stage, with everything but the 'graymail'
dumped by a simple recipe.

I prefer a straightforward procmail ruleset for that
stage.



[Note: I don't read the articles of "Sam" or his numerous
sockpuppets or his 'friends', nor any responses to them, and
haven't for years. He follows me all over the Usenet, and I
still don't read his articles. This _really_ pisses him off.
.]

Alan

--
http://home.earthlink.net/~alanconnor/contact.html
Other URLs of possible interest in my headers.

Re: Bayesian training [Was: Spam From Idiots]

This is a MIME GnuPG-signed message. If you see this text, it means that
your E-mail or Usenet software does not support MIME signed messages.
The Internet standard for MIME PGP messages, RFC 2015, was published in 1996.
To open this message correctly you will need to install E-mail or Usenet
software that supports modern Internet standards.

--=_mimegpg-commodore.email-scan.com-4703-1147258372-0003
Content-Type: text/plain; format=flowed; charset="US-ASCII"
Content-Disposition: inline
Content-Transfer-Encoding: 7bit

Usenet Beavis writes:

> On comp.mail.misc, in <20060510093552@usenet.piggo.com>, "Troy
> Piggins" wrote:
>
> Hey Troy!

Hey Beavis!

>
>> and I know you use a C/R system, but please indulge me.
>
> Sure.

You don't use a C/R system, Beavis.

>> So my question is this - should I pass those type of emails on
>> to SA's bayesian training system, or just delete them?
>
> Couldn't tell you, Troy,

But that won't stop you from flapping your gums, right Beavis?

> My hunch would be to delete them. If they are sent to throw off
> your filter, why give them a chance?

The golden rule of comp.mail.misc is: for every Beavis "hunch" or technical
advice, the correct answer always lies 180 degrees to the opposite.


> [Note: it's not my fault that I'm a complete dumbass. I was dropped on my
> head as a child. See http://www.pearlgates.net/nanae/kooks/ac/ for
> more information]
>
> Beavis



--=_mimegpg-commodore.email-scan.com-4703-1147258372-0003
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQBEYcYEx9p3GYHlUOIRAt0KAKCBN7QAAzsdlHfEet9O/ZDo7+PLdACf SERT
5dwDLdOW/gOph+w6O+R46FI=
=2bsq
-----END PGP SIGNATURE-----

--=_mimegpg-commodore.email-scan.com-4703-1147258372-0003--

Re: Bayesian training [Was: Spam From Idiots]

This is a MIME GnuPG-signed message. If you see this text, it means that
your E-mail or Usenet software does not support MIME signed messages.
The Internet standard for MIME PGP messages, RFC 2015, was published in 1996.
To open this message correctly you will need to install E-mail or Usenet
software that supports modern Internet standards.

--=_mimegpg-commodore.email-scan.com-4703-1147258556-0004
Content-Type: text/plain; format=flowed; charset="US-ASCII"
Content-Disposition: inline
Content-Transfer-Encoding: 7bit

Troy Piggins writes:

> I was hoping for someone with some knowledge about bayesian systems to
> reply. That's why I changed the subject heading. In hindsight maybe I
> should've started a new thread but I got lazy.
>
> Do you know anything about bayesian training in relation to my question?

It depends on the individual implementation.

Although something may seem random to you, there are patterns even to random
number generators.

A smart bayesian filter might pick up on the fact these spams -- for example
-- consist of two MIME sections, the text/html section containing a single
with an inside it that references a CID: of the second,
image/gif MIME section; as I suspect you'll find many of these spams
contain.

You can't really answer this without knowing the particulars of the bayesian
implementation.


--=_mimegpg-commodore.email-scan.com-4703-1147258556-0004
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQBEYca8x9p3GYHlUOIRAmIxAJ9/T+8LR2Sps+9L8fPJA9gM2uLAYwCf Tkwl
O9KP/p7tTrcFVwdc8VRZqWc=
=FwOt
-----END PGP SIGNATURE-----

--=_mimegpg-commodore.email-scan.com-4703-1147258556-0004--

Re: Bayesian training [Was: Spam From Idiots]

This is a MIME GnuPG-signed message. If you see this text, it means that
your E-mail or Usenet software does not support MIME signed messages.
The Internet standard for MIME PGP messages, RFC 2015, was published in 1996.
To open this message correctly you will need to install E-mail or Usenet
software that supports modern Internet standards.

--=_mimegpg-commodore.email-scan.com-4703-1147258652-0005
Content-Type: text/plain; format=flowed; charset="US-ASCII"
Content-Disposition: inline
Content-Transfer-Encoding: 7bit

Usenet Beavis writes:

> Postscript:
>
> On comp.mail.misc, in , "Usenet Beavis" wrote:
>
>>>
>>> I'm pretty sure you don't use SA.
>>
>> Correct.

How about a deodorant, Beavis? Do you use that?

> I should note that a lot of C/R users _do_ use SA as the
> conventional spam filter stage in their Challenge-Response
> system:
>
> http://www.geocities.com/suhatrasabib/

You don't say?

> [Note: it's not my fault that I'm a complete dumbass. I was dropped on my
> head as a child. See http://www.pearlgates.net/nanae/kooks/ac/ for
> more information]
>
> Beavis



--=_mimegpg-commodore.email-scan.com-4703-1147258652-0005
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQBEYcccx9p3GYHlUOIRAq4fAKCAFGSj1mRJtUmV3sT3syEDEDcVNgCf VwHV
l7wkX/Dwsnj3TXcnzAhCh0E=
=40kM
-----END PGP SIGNATURE-----

--=_mimegpg-commodore.email-scan.com-4703-1147258652-0005--

Re: Bayesian training [Was: Spam From Idiots]

* Alan Connor wrote:
> On comp.mail.misc, in <20060510093552@usenet.piggo.com>, "Troy
> Piggins" wrote:
>
> Hey Troy!

G'day Alan.

>> * Alan Connor wrote:
>>
>>> I have a mailbox reserved for spam, where the headers are
>>> stored for possible future study.
>>>
>>> wahoo77@yahoo.com
>>>
>>> Been noticing a lot of subjects like this:
>>>
>>> Subject: aardvark test oceans mega... [snip]
>> I think the above type of emails, at least the ones I have been
>> getting where the body is just a bunch of random phrases, are
>> meant to throw off learning/training type spam filters like
>> bayesian ones.
>
> Got it. Thanks.
>
>> I'm pretty sure you don't use SA.
>
> Correct.
>
>> and I know you use a C/R system, but please indulge me.
>
> Sure.
>
>> Those emails themselves are not spam (at least by my definition
>> since they are not selling anything, although they are
>> certainly unsolicited bulk emails).
>>
>> So my question is this - should I pass those type of emails on
>> to SA's bayesian training system, or just delete them?
>
> Couldn't tell you, Troy, but someone will.
>
> My hunch would be to delete them. If they are sent to throw off
> your filter, why give them a chance?

Based on some advice from others here, and also some reading I'm doing
(or rather started), SA also learns from headers so it is probably
worthwhile using them for training as well.

Peace.

--
Troy Piggins RLU#415538
My dogma just ran over my karma

Re: Bayesian training [Was: Spam From Idiots]

* Sam wrote:
> Troy Piggins writes:
>
>> I was hoping for someone with some knowledge about bayesian systems to
>> reply. That's why I changed the subject heading. In hindsight maybe I
>> should've started a new thread but I got lazy.
>>
>> Do you know anything about bayesian training in relation to my question?
>
> It depends on the individual implementation.
>
> Although something may seem random to you, there are patterns even to random
> number generators.
>
> A smart bayesian filter might pick up on the fact these spams -- for example
> -- consist of two MIME sections, the text/html section containing a single
> with an inside it that references a CID: of the second,
> image/gif MIME section; as I suspect you'll find many of these spams
> contain.
>
> You can't really answer this without knowing the particulars of the bayesian
> implementation.

Fair enough. Thanks for the pointers. I'll do some more reading
hopefully this weekend.

Peace.

--
Troy Piggins RLU#415538
My dogma just ran over my karma