About Bayes engine
Desai, Jason
jase at SENSIS.COM
Wed May 12 19:19:18 IST 2004
Carl,
I've attached my perl script. It currently looks for spams with a score > 7
which have not already been auto learned, and feeds them to sa-learn. You
will want to adjust some variables at the top for your setup.
Jase
Carl Boberg wrote:
> Jase,
> I would be very happy if I could take a look at your script.
> I could really use it. I know I should make my own but why
> invent the wheel twice :-) Besides, Im short on time and
> not so good at scripting :-(
>
> / Carl
>
>> -----Original Message-----
>> From: MailScanner mailing list [mailto:MAILSCANNER at JISCMAIL.AC.UK]On
>> Behalf Of Desai, Jason Sent: den 12 maj 2004 14:49
>> To: MAILSCANNER at JISCMAIL.AC.UK
>> Subject: Re: About Bayes engine
>>
>>
>> If you're using MailWatch, you could write a simple script to
>> query the
>> database, grab quarantined files that match your search
>> criteria, and feed
>> them to sa-learn. I've done something similar and it works fine.
>>
>> Jase
>>
>> kfliong wrote:
>>> Hi,
>>>
>>> I am using manual Bayes. That means I feed bayes with spams and hams
>>> manually because my server gets like 26539 mails per day. After
>>> feeding it 200 spams and hams, it started to work yesterday. So now
>>> I am getting from around 88% spams previsouly to 96% spams. This is
>>> good. But I know I have to keep on feeding bayes with spams in order
>>> to increase it's accuraccy. I don't mind feeding it like 10-20 mails
>>> a day. But is this method effective? Do you see any problems with
>>> this method?
>>>
>>> I also saw something about Rebuild Bayes Every option. Does that
>>> means the bayes will delete the current learned mails and relearn
>>> them? Of course this would only be applicable to bayes that
>>> auto_learn.
>>>
>>>
>>> thanks
>>>
>>> -------------------------- MailScanner list ----------------------
>>> To leave, send leave mailscanner to jiscmail at jiscmail.ac.uk
>>> Before posting, please see the Most Asked Questions at
>>> http://www.mailscanner.biz/maq/ and the archives at
>>> http://www.jiscmail.ac.uk/lists/mailscanner.html
>>
>> -------------------------- MailScanner list ----------------------
>> To leave, send leave mailscanner to jiscmail at jiscmail.ac.uk
>> Before posting, please see the Most Asked Questions at
>> http://www.mailscanner.biz/maq/ and the archives at
>> http://www.jiscmail.ac.uk/lists/mailscanner.html
>>
>
> -------------------------- MailScanner list ----------------------
> To leave, send leave mailscanner to jiscmail at jiscmail.ac.uk
> Before posting, please see the Most Asked Questions at
> http://www.mailscanner.biz/maq/ and the archives at
> http://www.jiscmail.ac.uk/lists/mailscanner.html
-------------------------- MailScanner list ----------------------
To leave, send leave mailscanner to jiscmail at jiscmail.ac.uk
Before posting, please see the Most Asked Questions at
http://www.mailscanner.biz/maq/ and the archives at
http://www.jiscmail.ac.uk/lists/mailscanner.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: learnspam
Type: application/octet-stream
Size: 3006 bytes
Desc: not available
Url : http://lists.mailscanner.info/pipermail/mailscanner/attachments/20040512/325acac4/learnspam.obj
More information about the MailScanner
mailing list