<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Tony Meyer wrote:
<blockquote TYPE=CITE>1. SpamBayes doesn't use any tokens that have
a current spamprob between
<br>0.4 and 0.6 (you can change these values if you like). So 0.62
is just
<br>outside that range, and so it does appear to have a little bit of value
<br>(indicating that mail is just a wee bit more likely to be spam).</blockquote>
OK, that makes sense. So, (other than ignoring the problem ;-) I could
either move the "goalposts", or find some ham that came through that mail
gateway and do some more training.
<br>
<blockquote TYPE=CITE>it's
<br>basically doing what you've asked, but automatically, rather than via
some
<br>manual tool.</blockquote>
That's cool.
<p>I wasn't clear on how the classifier selected its evidence (nor how
the individual terms are weighted).
<br>
<blockquote TYPE=CITE>2. The 150 'strongest' (furtherest from 0.5)
tokens are used, by default.
<br>Early testing showed that this was a good number, but if you like,
you can
<br>change this, too - if you set it high enough, then every token will
be used,
<br>no matter what score it has.</blockquote>
Well, I only counted about 80 in the mail header, but, uh, I wasn't exactly
counting carefully. Perhaps my training corpus was too small to complete
cover this piece of spam?
<p>Anyway, I was wondering why there's no hand-tuning option, and I think
you answered the question. So, I'm just going to ignore the evidence
(which is how any good logician proceeds ;-).
<br>
<p> Thanks,
<p>
Webb
<br>
<pre>--
------------------------------------------------------------------------
Webb Scales Hewlett-Packard Company
scales@zko.dec.com 110 Spit Brook Rd, ZKO2-3/N30
Voice: 603.884.2196, FAX: 603.884.0120 Nashua, NH 03062-2711
Someone who thinks logically provides a nice contrast to the real world.
------------------------------------------------------------------------</pre>
</html>