home | books | articles | gleanings | case studies | hire
other sites: widgetopia | blueprints for the web | metafooder | Mammahood


 


 


« getting excited | main | go learn something »

trying to come up with a strategy

I'm hurting with spam. I'm looking for help.

I foudnthis facinating descriptio fo how one tool works: SpamBayes: Bayesian anti-spam classifier written in Python.

"The system then uses these clues to examine new messages.
For instance, the word "Nigeria" appears often in spam, so you could use a spam filter which identifies anything with that word in it as spam. But what if your business involves writing a guidebook on Nigerian Wildlife Conservation? Clearly a more flexible approach is necessary. Additionally spammers will adapt their content over time and will no longer use the word "Nigeria" (or the words "Lose Weight Fast", or any number of other common lines). Ideally the software will be able to adapt as the spam changes.
So, that is what SpamBayes does. It compares the spam and the ham and calculates probabilities. "

Posted at January 24, 2004 01:05 PM


Comments

 

I would recommend that you look into POPFile http://popfile.sourceforge.net/

POPFile also works as an bayesian filter which you have to train to classify your mail. You install it locally, which is very easy, and then after fairly little training you have an effective spamfilter. I currently have a successrate of 99%, so it is rather effective! :-)

Posted by Carsten Holst at January 26, 2004 3:16 PM


~~~



Post a comment
*Name:


*Email Address:


URL:


Remember me?

Comments:

bold italic underline link


posting can be slow; please wait a few seconds before hitting the button again.

The extra-fine print
wording stolen by the more-eloquent-than-I kottke
The bold, italics, and link buttons (and associated shortcut keys) only work in IE 5+ on the PC.
Hearty discussion and unpopular viewpoints are welcome, but please keep comments on-topic and *civil*. Flaming, trolling, and ass-kissing comments are discouraged and may be deleted.
All comments, suggestions, bug reports, etc. related to the comments system should be directed to me.


mail entry to a friend

Email this entry to:


Your email address:


Message (optional):




« getting excited | main | go learn something »

 

 

 

home | books | articles | gleanings | case studies | hire
other sites: widgetopia | blueprints for the web | metafooder | Mammahood