Bayesian Filter
September 3, 2018
An Email Marketing Glossary Entry

What is a Bayesian Filter?

Bayesian filters are adaptive programs that help to identify incoming email traffic as spam by calculating probabilities based on the message's contents.

How Bayesian Spam Filtering Works

A well-tuned Bayesian spam filtering program can learn from spam and from good mail, and return few if any false positives. You start by feeding the filtering program with two sets of emails: one with messages you have classified as spam, and the with valid emails. The filters look at both, analyzes the legitimate mail and the spam, and calculate the probability of various characteristics appearing in spam versus in a good mail.

Spam will (almost) always differ from regular emails. The side effect in trusting such anti-spam filters is that errors — which cannot ruled out — can have significant effects, either through false positive or false negatives.


The Mailman

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram