Does Bayesian poisoning exist?

2006-02-01

John Graham-Cumming

The POPFile Project, France
Editor: Helen Martin

Abstract

A common criticism of statistical spam filters (such as SpamBayes, DSPAM and POPFile) is that they can be 'poisoned' by inserting random words into spam messages. John Graham-Cumming looks at the evidence and attempts to answer the question: does Bayesian poisoning exist?

Copyright © 2006 Virus Bulletin

The full article is available to registered users. Click here for free registration or, if you already are a registered user, login to access the full article.

Poll

How should software and OS patching/security updates be managed?
Manually, at the user's discretion
Automatically via an optional, user-defined schedule
Automatically via a fixed, but optional schedule
Automatically via a fixed schedule, on by default with opt-out system
Automatically and silently, with no option to run unpatched

Leave a comment
View 19 comments
Jobs Career Sidebar

Virus Bulletin

In this month's magazine:
  • Same malware, different code
  • Can you spare a seg?
  • Kernel mechanics of Rustock
  • Early warning approaches to combat typosquatting
  • The challenges of collecting and monitoring URLs that point to malware
  • Norman Network Protection Appliance
  • Anti-spam comparative review July 2009
Virus Bulletin 07 2009
Subscribe now!
Virus Bulletin currently has 165,654 registered users.