Hi folks,
sorry for asking again but Bayes is a blackbox for me :(
I have some question about bayes math. How does bayes/spamassassin
calculate this values?
i'm wonder about : auto-learn?* and "score so far=8.601)
dbg: rules: running head tests; score so far=8.601
[...]
dbg: rules: running body tests; score so far=8.601
[...]
dbg: rules: running uri tests; score so far=8.601
[...]
dbg: rules: running body_eval tests; score so far=8.601
[..]
dbg: rules: running rawbody tests; score so far=8.601
[...]
dbg: rules: running full tests; score so far=8.601
[...]
dbg: rules: running meta tests; score so far=8.601
[...]
dbg: learn: auto-learn: currently using scoreset 3, recomputing score
based on scoreset 1
dbg: learn: auto-learn: adding body_only points 0.001
dbg: learn: auto-learn: not considered head or body scores: 0.1
dbg: learn: auto-learn: adding head_only points 1.274
dbg: learn: auto-learn: adding head_only points 0.01
dbg: learn: auto-learn: message score: 9.404, computed score for
autolearn: 1.385
dbg: learn: auto-learn? ham=-1, spam=6.5, body-points=0.001,
head-points=1.284, learned-points=8.5
can someone explain this? ore give me some reading stuff? i can't find
anything else...
PS:
I have already read the following
http://www.paulgraham.com/naivebayes.html(like spamassassin say it use
bayes like this)
http://www.paulgraham.com/spam.html
https://cwiki.apache.org
Kind regards
--
Philipp Ewald
Administrator