re-traning Bayes. i made a boo-boo

General Interworx Control Panel Info General Inquries
1

hi all:

while moving some messages around in Horde, i accidentally moved already-tagged messages (i.e. [SPAM]) into the “Learn Spam” folder. now, all tagged messages have been learned and come to the inbox with double tags (see below and notice the double tagging in paragraphs 1 and 2). how do i undo this boo-boo?

Spam detection software, running on the system “host.server1.com”, has
identified this incoming email as possible spam. The original message
has been attached to this so you can view it (if it isn’t spam) or label
similar future email. If you have any questions, see
the administrator of that system for details.

Content preview: Spam detection software, running on the system “host.server1.com”,
has identified this incoming email as possible spam. The original message
has been attached to this so you can view it (if it isn’t spam) or label
similar future email. If you have any questions, see the administrator of
that system for details. […]

Content analysis details: (12.0 points, 4.7 required)

pts rule name description

0.5 DATE_IN_PAST_03_06 Date: is 3 to 6 hours before Received: date
0.0 UNPARSEABLE_RELAY Informational: message has unparseable relay lines
0.1 OPTING_OUT_CAPS BODY: Talks about opting out (capitalized version)
1.9 EXCUSE_24 BODY: Claims you wanted this ad
0.0 HTML_MESSAGE BODY: HTML included in message
3.5 BAYES_99 BODY: Bayesian spam probability is 99 to 100%
[score: 1.0000]
3.0 URIBL_BLACK Contains an URL listed in the URIBL blacklist
[URIs: bedtowel.com]
3.0 URIBL_OB_SURBL Contains an URL listed in the OB SURBL blocklist
[URIs: bedtowel.com]

The original message was not completely plain text, and may be unsafe to
open with some email clients; in particular, it may contain a virus,
or confirm that your address can receive spam. If you wish to view
it, it may be safer to save it to a file and open it with an editor.

i exported the bayes database in hopes to edit it and then re-upload it, but, suffice it to say, isn’t possible. anyone have this before? i basically need to undo some of the training…

regards,
j

2

after clearing the db, i started over…had about 1500 ham/spam messages, then monitored the performance (hits). after all this, i decided to ditch bayes.

conclusion…don’t use bayes. most of the spammers are using some form of poisoning anyway by including “acceptable” text at the bottom of their emails, which further makes bayes less effective. also, bayes is a resource hog.

john

3

…had about 1500 ham/spam messages, then monitored the performance (hits). after all this, i decided to ditch bayes.

conclusion…don’t use bayes. most of the spammers are using some form of poisoning anyway by including “acceptable” text at the bottom of their emails, which further makes bayes less effective.

Read the post on this thread? (or your’s were image spams?)

http://interworx.com/forums/showthread.php?t=2604 (mainly post #1: http://interworx.com/forums/showpost.php?p=14734&postcount=1)

4

yes, i did read this thread. good info. to answer your question tiger…no, mostly they were text.