Vincenzo,

Thank you so much for explanation and the offer.  Yes, please do send me the
mySQL dump of your corpus.  That would give me a big headstart.

Thanks again.

Jerry

-----Original Message-----
From: Vincenzo Gianferrari Pini [mailto:[EMAIL PROTECTED] 
Sent: Wednesday, January 10, 2007 4:14 AM
To: James Users List
Subject: Re: Best way to feed ham or spam to BayesianAnalysis mailet?

Hi JWM,

I'm happy to see that the mailet is useful.

The reason why a "real" attachment is needed is because the feeder 
strips out all the headers of the mail you send it, but considers as 
spam/nospam the headers in the attached message, which is the original 
one. This is because you don't want your own name/address/subject and 
other header stuff coming from your smtp client to be considered spam 
related tokens.

BTW mail header info is *very* important to the spam bayesian 
recognition, so you definitely want the original mail header info to be 
analyzed.

With Outlook, you simply create a new empty message with the correct TO 
recipient, drag+drop the original message into it and send it. It will 
become a new message with the original as attachment, which is different 
from a forward of the original.

Vincenzo

P.S. If you want, I can send you off-list a mysql dump of my working 
corpus, that has been fed with thousands of spams/nospams. Just let me know.


JWM wrote:
> OK, dumb question... I understand what an attachment is in email, etc.
But
> I never really considered a 'forward' of an email to be sending the
original
> as a attachment.
>
> I'm a plain old "Outlook" email user. If I get a spam and simply hit the
> 'forward' button and send it to the spam/ham email address, is that going
to
> work?  Or do I need to do some other gyrations with saving it to a file
and
> dragging the file to a new note kinda like I'd do with a jpg?
>
> In other words, how do I send ham/spam using Outlook?  What is the
> significance of it being an attachment?
>
> Thanks.
>
> Jerry
>
> -----Original Message-----
> From: Norman Maurer [mailto:[EMAIL PROTECTED] 
> Sent: Monday, January 08, 2007 3:57 AM
> To: James Users List
> Subject: Re: Best way to feed ham or spam to BayesianAnalysis mailet?
>
> Danny Angus schrieb:
>   
>>> I've only added a relatively small corpus so far and the Bayesian filter
>>> is doing a fantastic job.
>>>       
>> It is pretty awesome isn't it?
>> Respect to Vincenzo for his efforts.
>>
>> d.
>>
>>     
> That was why i asked for a subproject ;-)
>
> bye
> Norman
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>
>   


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to