On Fri, Jul 26, 2019 at 04:50:21PM -0600, Amir Caspi wrote:
>
> If only the first spam had come 15 seconds later, Razor would have caught it.
No, it could be 15 seconds or 15 minutes later or never. Razor is not a
magic de-poisoner, it still calculates different hashes for these mails.
On Jul 26, 2019, at 4:24 PM, RW wrote:
>
> Most of the difference in score came from RAZOR2_CHECK and
> RAZOR2_CF_RANGE_51_100 rather than FSL_BULK_SIG.
Thanks to both you and Bill for the clarification. If only the first spam had
come 15 seconds later, Razor would have caught it. What a dif
On 26 Jul 2019, at 16:52, Amir Caspi wrote:
Hi all,
In recent weeks I've been receiving many of my spams in doubles --
essentially identical spam except for the faked From and the various
"Bayes poison" random text. I just got one such pair where
FSL_BULK_SIG fired on one spam, but not the
On Fri, 26 Jul 2019 14:52:11 -0600
Amir Caspi wrote:
> Hi all,
>
> In recent weeks I've been receiving many of my spams in doubles --
> essentially identical spam except for the faked From and the various
> "Bayes poison" random text. I just got one such pair where
> FSL_BULK_SIG fired on one sp
Hi all,
In recent weeks I've been receiving many of my spams in doubles -- essentially
identical spam except for the faked From and the various "Bayes poison" random
text. I just got one such pair where FSL_BULK_SIG fired on one spam, but not
the other, even though their content (except for th
I am a Google summer of code student working on developing a statistical
classifier plugin.
I have my models ready and started the integration part. As of now, I'm
taking the Bayes.pm file as the baseline. Any other documentation which I
can follow or a plugin which I can take as a reference? ( I
