August 29, 2002
Statistical analysis of spam

Lars is looking into statistical analysis of spam. He has a corpus of some 2000 messages filtered by Spam Assassin the last couple of months, and I am sharing data for my own corpus of similar size.

Other similar investigations & rants:

(source: Corante: The Silver Bullet to Kill Spam)

Update: Lars has published the first results of his investigations.

Sponsored links
Related Entries

According to discusstions on the SpamAssassin list, it appears that the next release of SpamAssassin (2.4) will have Bayesian filtering.

Posted by: Adam Kalsey on August 30, 2002 07:38 AM

SpamWeed is a very effective anti-spam filter.

Posted by: L.B.S. on May 22, 2003 02:29 PM
Post a comment

(Will not be displayed if you enter a website below. Otherwise, it will be displayed "spam protected")

(if you have one)

What do you want to say?
(please don't bother posting "spam" (pornography, viagra-sales etc - I will delete such comments anyway))

Remember info?

Referrers to this page
TrackBack URL for this entry:
myBlog by Lars: Statistical Analysis of Spam - Part I (August 29, 2002 09:11 PM)
"As Anders pointed out earlier today I have been working on analysing spam statistically. I made a brief introduction to"
myBlog by Lars: Statistical Analysis of Spam - Part II (September 2, 2002 05:51 PM)
"Just after finishing my last piece on this topic (Statistical Analysis of Spam - Part I) I recieved a list of some 1900 X-Spam-Status lines from Anders. Inserting this into Minitab I expected to get two graphs looking more or less similar to eachother...."
myBlog by Lars: Statistical Analysis of Spam - Part III (September 3, 2002 07:10 PM)
"Once more it is time to study the statistical aspects of spam. In part II I commented on the differences in two independent sets of spam. Today I will try to be slightly more scientific. Just by looking at the histograms in the previous posting it is e..."

[an error occurred while processing this directive]

© Anders Jacobsen
[ photography]