Bayesian filtering question.

Protecting your Linux box
azfar
Captain
Posts: 598
Joined: Tue Mar 23, 2004 1:16 am
Location: Karachi
Contact:

Bayesian filtering question.

Postby azfar » Sat Nov 29, 2008 3:01 pm

I am trying to find out how exactly bayesian filtering works but I am not clear on it.

My question is how bayesian filtering actualy works i-e it collects the words from whole message (header, body & subject) or just body & subject or just from body.

In my scenario users forward me those spam messages (ham) which are incorrectly marked as spam and those ham messages (spam) which are actualy spam.

My problem in this way is that orignal messages are now modified and extra headers and signatures are also added with them and I am curious that this will train the filter in wrong manner.
Azfar Hashmi
Email : azfarhashmi@hotmail.com

lambda
Major General
Posts: 3452
Joined: Tue May 27, 2003 7:04 pm
Location: Lahore
Contact:

Postby lambda » Sat Nov 29, 2008 4:48 pm

Watch out for the !
Isn't it amazing how so many people can type "linuxpakistan.net" into their browsers but not "google.com"?

azfar
Captain
Posts: 598
Joined: Tue Mar 23, 2004 1:16 am
Location: Karachi
Contact:

Postby azfar » Sat Nov 29, 2008 9:32 pm

Azfar Hashmi

Email : azfarhashmi@hotmail.com


Return to “Security”

Who is online

Users browsing this forum: No registered users and 1 guest