// COMPLETE THREAD

Textual Analysis of Spam

1 expanded post ยท every known parent and child

NODE 17ff1963Textual Analysis of Spam
I'm currently writing a spam-filter and bouncing messages based on header
information and keyword counts. This gets about 80% of the spam, but also
gives about 1% false positives. Has anyone done any more sophisticated
analysis of spam text than this? I can't find anything on the Web.

	Mark