Re: [exim] Testing an interesting spam catching theory

Pàgina inicial
Delete this message
Reply to this message
Autor: Marc Perkel
Data:  
A: . kibble .
CC: exim-users
Assumpte: Re: [exim] Testing an interesting spam catching theory
thanks - I've started testing this and the results are very interesting
so far. The accuracy is good. It's still training though.

. kibble . wrote:

>
>> On: Fri, 14 Jan 2005 17:53:52 Marc Perkel Wrote:
>>
>> I have a theory that if a person ran a bayesian filter only on the
>> headers that it would be more accurate that running it on the body of
>> the message as well. And - I'm in the process of trying to test it
>> using spamprobe but feeding spamprobe only headers. I'm using Spam
>> Assassin to select the headers that it will auto train on.
>>
>> If it works - I will share with all.
>>
>
> I have a piece of code [compiled in with the local_scan API] which
> depends on MySQL, DSPAM and ClamAV, although currently it scans the
> entire mail message [in mbox format], I can quiet easily adapt it to
> scan either the whole message, headers only, or body only…
>
> Not sure if that would be of any interest to you, or if the
> complications of compiling local_scan and the extra ‘configurational’
> [<- if that’s such a word] changes are worth the overall out come…
> Maybe useful to those who are starting a fresh and want virtual users
> in a MySQL database would find this of some use ?
>
> _________________________________________________________________
> It's fast, it's easy and it's free. Get MSN Messenger today!
> http://www.msn.co.uk/messenger
>
>