Kevin wrote:
> Patrick Butts wrote:
>   
>> I've been rebuilding twice a day since day one.. (it's in a cron) and 
>> I've ran the move2num.pl twice in two weeks.  
>>     
>
> Unless you have/had 'UseSubjectsAsMaillogNames' enabled there is no 
> reason to run move2num.pl.
>   
I was using the Subjects.  I quit today and ran the move2num for the 
second and last time.
> The move2num.pl script is used to rename corpus emails from their 
> subject to numbers, the naming is required for ASSP to manage the size 
> of the corpus.
>   
Aye...

>> I'm guessing my biggest problem is that I have 5500 spam in the folder
>> and only 682 notspam in the other folder. 
>>     
>
> Yes, that is the problem. Disable the Bayesian scanning until your 
> corpus has time to build up, don't simply put it in testmode as that 
> will only make your corpus even worse.
>   
Done.  I also set freqSpam:=9 instead of 1.  I figure I have about 10% 
nonspam to spam so I'll save less spam until it starts to catch up.


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Assp-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/assp-user

Reply via email to