bogofilter resistant email
Pavel Kankovsky
peak at argo.troja.mff.cuni.cz
Sun Feb 22 18:49:54 CET 2004
On 16 Feb 2004, Tom Anderson wrote:
> The end result will be that words such as "the", "these", "their", etc.,
> will be considered spammy.
IMHO, the end result should probably be to make those words neutral from
the classificator's pov. After all, there are the typical "stop words".
> And if that is the case, then I must prepare to receive lots of false
> positives.
I myself would do the transition in multiple gradual steps, watch ham
scores carefully, and train more hams as needed. Moreover, it is always a
good idea to test a substiantially modified db against a corpus of known
spam & ham before you deploy it.
--Pavel Kankovsky aka Peak [ Boycott Microsoft--http://www.vcnet.com/bms ]
"Resistance is futile. Open your source code and prepare for assimilation."
More information about the Bogofilter
mailing list