bogotune results
Tom Allison
tallison at tacocat.net
Wed Mar 24 12:27:26 CET 2004
Boris 'pi' Piwinger wrote:
> Tom Allison <tallison at tacocat.net> wrote:
>
>
>>Well, it took a while, but this is what bogotune finally spit out as the
>>results.
>>These are seom pretty weird numbers...
>>I retested my archive and of the Ham I got: 3 Unsure and 1 Spam
>>of the Spam I got: all spam.
>
>
> Was this before or after running bogotune?
After.
Before my test results in 100.00% scores.
>
>
>>The highest fpos (Spam) was 0.238562. So I guess it's safe to set that
>>as my spam_cutoff for now.
>
>
> Also: This is after changing the parameters?
Yes, previously was all default except robx=0.60
>
>
>>tallison at janus:~> bogotune -n Maildir/.training.ham/cur/ -s
>>Maildir/.training.spam/cur/
>>Calculating initial x value...
>>Initial x value is 0.600000
>>Too few high-scoring non-spams in this data set.
>>At target 1, cutoff is 0.238562.
>
>
> This here suggests that this is after choosing the initial
> values by bogotune.
>
>
>>Recommendations:
>>
>>---cut---
>>db_cachesize=4
>>robx=0.600000
>>min_dev=0.020
>>robs=0.0100
>>spam_cutoff=0.069 # for 0.05% fpos (1); expect 0.00% fneg (0).
>>#spam_cutoff=0.040 # for 0.10% fpos (2); expect 0.00% fneg (0).
>>#spam_cutoff=0.020 # for 0.20% fpos (4); expect 0.00% fneg (0).
>>ham_cutoff=0.020
>>---cut---
>
>
> So why does bogotune suggest those values? Zero false
> positives were easily possible with a higher ham_cutoff.
>
> The new values seem to give much lower values for your
> problematic messages.
>
Yeah, I guess I can keep bumping up the cutoffs.
I set ham_cutoff=0.24 and spam_cutoff 0.4 for now.
More information about the Bogofilter
mailing list