Need some help to improve on Spam filtering.

My ISP is has a spam filter. Based on their default setting I lost to much
mail as spam. I changed it so now 10% of my mail is spam. Which is OK since
I use BayesIt as an internal filter.

The problem is that BayesIt designate to many mails as non-spam.(Spam error
is 20%). I cannot remember any mail which was designated as spam by BayesIt
in error.

Can I change some setting in the advance.ini file or somewhere else to
improve the filtering of Spam mail?

My advance.ini file and information (abbreviated) is below

I haven't much clue what the lines in advance.ini do and therefore do not
want to experiment with it.

Can anyone suggest a change in any settings to improve filtration?


Advance.ini file:

> working thread priority="2"
> onexit thread priority="3"
> selective download spam threshold="10"
> export selective download="1"
> simple digits spam marks="1"
> no spaces spam marks="1"
> limit size to hash="19"
> limit size to hash header="96"
> temporary dictionary="c:\\temp"
> use expiration="0"
> age to expirate="100"
> learn from zero="1"
> max size of log file="131072"
> recalculating strategy="3"
> regarding threshold="1.5"
> use autotrain="1"
> use degeneration="1"
> number of exclamations="5"

 Plug-in Information:

> Antispam filtering data:
> 
> Spam frequency dictionary:
> °     Size: 907 letters.
> °     Capacity: 70713 words.
> Non-spam frequency dictionary:
> °     Size: 5397 letters.
> °     Capacity: 127363 words.
> Current active base:
> °     Active current base contains 32992 words.
> °     Status: OK
> 
> Last month statistic
> General numbers
> °     Spam traffic (bytes): 2583757
> °     Spam letters: 337
> °     NON-spam traffic (bytes): 17897140
> °     NON-spam letters: 3028
> °     Total traffic (bytes): 20480897
> °     Total letters: 3365
> °     Part of the spam in terms of letters: 10.01%.
> °     
> Errors
> °     SPAM errors (letters): 24.04%.
> °     SPAM errors (traffic): 20.09%.
> °     NON-spam errors (letters): 0%.
> °     NON-spam errors (traffic): 0%.
> °     Totally errors (letters): 2.41%.
> °     Totally errors (traffic): 2.53%.
> °     
> Current running version is 0.5.9

-- 
Best regards,
Peter Kerekes (Toronto, Canada) 

"If you are never scared, embarrassed or hurt, it means you never take chances." 
      -- Julia Soul   

TB! v2.12.03 and BayesIt! 0.5.9 on Windows 20005.0.2195 Service Pack 4


________________________________________________
Current version is 2.12.00 | 'Using TBUDL' information:
http://www.silverstones.com/thebat/TBUDLInfo.html

Reply via email to