Print Page | Close Window

Bayesian filter

Printed From: LogSat Software
Category: Spam Filter ISP
Forum Name: Spam Filter ISP Support
Forum Description: General support for Spam Filter ISP
URL: http://www.logsat.com/spamfilter/forums/forum_posts.asp?TID=5355
Printed Date: 19 October 2017 at 1:19am


Topic: Bayesian filter
Posted By: Guests
Subject: Bayesian filter
Date Posted: 11 October 2005 at 4:49am

Hi.
The filter has only kicked in a very few times. Last time I checked Corpus.ini 1.500.000 SPAMS and 900.00 GOOD mails had passed the filter. I deleted the Corpus-folder to "restart" it.

When I dump the Corpus I found that the Probabilityfigure is written with the Norwegian standard, using , (comma) and not . (period) as the decimaldelimiter. Is this a problem or will SPAMfilter handle this?

*Token ,Good, Spam, ProbSpam, ModDate
*!&nbsp,1,0,0,400000005960464,2005-10-11
*!jx,0,1,0,400000005960464,2005-10-11
*$0*12,1,1,0,400000005960464,2005-10-11

 

Regards
Eilif Skappel




Replies:
Posted By: eilif
Date Posted: 11 October 2005 at 8:21am

I forgot to tell you that I'm using v.2.6.3.487 on W2K.

Regards
Eilif Skappel



Posted By: LogSat
Date Posted: 11 October 2005 at 11:11pm
Eilif,

The "dump corpus" does a very basic memory dump of the corpus database to a text format, using a hard-coded comma to separate the fields. It's used for troubleshooting purposes only.

Internally, the numeric data is stored in native, binary format (not in text mode), and the same applies when storing it in the corpus database file. As such, there should not be any issues with the locale being used.


-------------
Roberto Franceschetti

http://www.logsat.com" rel="nofollow - LogSat Software

http://www.logsat.com/sfi-spam-filter.asp" rel="nofollow - Spam Filter ISP



Print Page | Close Window