Bayesian Spam Filters Explained
In a single word Bayesian spam filters are "intelligent". Bayesian spam filters are intelligent in so far as they’re capable of comparing two sets of information and acting on the result. This is in direct contrast to the vast majority of other spam filters who use pre-built rules to decide which email is spam and which is not.
Bayesian spam filters can take one group of legitimate email and another group of spam and compare the values and data of each. The definition of legitimate email that it creates at the end of this comparison session is what it uses going forward to scan your inbox for spam.
For your information, Bayesian spam filters are named after Thomas Bayes an 18 century cleric who created something known as Bayes Theorem. In summary Bayes Theorem is as follows: "in statistical inference to update estimates of the probability that different hypotheses are true, based on observations and a knowledge of how likely those observations are, given each hypothesis." In plain English it looks for obvious repeating patterns to form an "opinion" on something. In spam filter terms that "opinion" becomes a rule which keeps you spam free (or pretty close.)
The really neat thing about Bayesian filters is that they’re capable of learning. For example if they decided to block an email because the filter perceived it as junk but the user marked it as valid mail the Bayesian filter then knows not to block that type of email in the future. So, in time, this type of spam filter learns enough to block spam far more effectively. AOL have embraced this type of spam filter with the launch of AOL 9.0 and AOL Communicator- if the big dog wants it then it must be worthwhile.
So what Bayesian spam filtering options are available to you? Well quite a few to be honest and you’ll be pleasantly surprised by some of the names involved. The first one on the list is AOL with their AOL Communicator product. The spam filtering features in AOL Communicator and AOL 9 are, to be honest, impressive. Think what you will of the provider themselves AOL Communicator is an excellent product and is suitable for use by both PC and Mac OSX users.
Next up we have Eudora. The nice folks at Qualcomm have designed an excellent email client that also has built in Bayesian spam filtering. I’ve used Eudora in the past and it’s a neat little package. Again the benefits here are advanced integrated spam filtering with your e-mail automatically. Mac OSX and OS9 users are in luck with Eudora providing support for both.
Bayesian spam filters are providing a more rounded defense for computer users of all levels in fighting spam. As an indication of the strength of Bayesian spam filters more and more companies are either:
- Updating their spam filters to include bayesian technique as a plugin or;
- Basing their new products on bayesian filtering principles from the ground up.