Results skewed by headers

[RW]

On Fri, 15 May 2009 10:50:00 +0930
Stephen Davies <scldad at sdc.com.au> wrote:


> The first five apparently outweigh the negative results.
> As soon as I run this through bogofilter -Ns, it becomes recognised
> as spam. This seems to confirm that the from and url headers are the
> most significant.
> 
> Why is this and what can I do to stop it happening in future?

It looks to me as if you miss-trained on a similar spam. The first 4
tokens are very specific and have a count of 1.