Outlook Express - How To Filter Spam (Part Two of Two)

Outclass

Outclass integrates POPFile's classification engine into Outlook.

Pros:

* You get an easy to use toolbar that lets you re-classify messages.

* You can set up "buckets" (or categories) easily and have Outclass file or delete matching messages automatically.

* Outclass works with IMAP or Exchange accounts just like incoming POP mail.

Cons:

* Some users report instability and crashing under some circumstances.

* Outclass doesn't work neatly with Outlook's vertical pane layout.

SpamAid

Using Bayesian statistics from a database of existing spam, no initial training period is necessary.

Unfortunately you will have to rely on someone else's definition of spam in the beginning so you may see some false positives right out of the box.

In addition to Bayesian filtering, SpamAid uses "blacklists" and "whitelists" to streamline spam filtering. It also uses a system of "phrase filters" that you can customize.

Pros

* Uses Bayesian filtering with a high accuracy

* You can "spam-test" your own outgoing mail for its potential of being caught in someone else's spam filter

* Easy to use

Cons

* You cannot tell SpamAid to scan certain folders automatically.

* False positives can be expected when you start using it.

SpamBayes Outlook Addin

The SpamBayes Outlook Addin leverages the proven track record of Bayesian spam filtering technology.

Pros:

* SpamBayes can show you a detailed analysis of a message's overall spam probability.

Cons:

* Not easy to sub-categorize "good email" into separate folders

SpamCatcher Outlook

On the plus side, SpamCatcher combines a content-scoring filters with a user network that is providing feedback on their spam. It can be used as a POP proxy so you can use it with most email clients, not just Outlook.

On the negative side, you can't easily customize the filters are not customizable. This might be the reason why it returns a significant number of false positives (and negatives). Not only that: you can't submit corrections to the user network with SpamCatcher.

Pros

* Easy to use

* Spam classified by the user community network as well as Bayesian analysis

* Includes automatic filter updates

Cons

* Not the most precise filters

* Cannot correct false positives

* Only works with POP accounts, not IMAP

Trend Micro Anti-Spam Pilot

Anti-Spam Pilot uses heuristic filters to spot and isolate spam. This is a somewhat older technology where you filter the email by looking for patterns that might match a collection of known spam. For example, the anti spam software might find certain words, phrases, punctuation, that give it a clue that this is spam in your email box. It then provides a score for the email. If the score is too high, the anti spam software will isolate the spam into a separate folder or do anything else you specify.

This approach is different than a Bayesian filter that actually learns as you go. The upside of this is that you don't have to spend a lot of time "training" the filter. The downside is that it doesn't detect as much spam as some other packages using Bayesian technology.

You do have the option of reporting missed spam (false negatives) and the rare spam emails that escape through the filter (false positives). You can also keep a list of spammers (a blacklist) and a list of legitimate emailers (a whitelist) as a way to increase Anti-Spam Pilot's overall effectiveness.

Pros

* Simple and uncomplicated to use

* Detects a good deal of spam

* Low number of false positives

* It's easy to reclassify email mistaken as spam (and vice versa)

Cons

* Does not filter IMAP accounts, only POP accounts

* Only works with Outlook

[Note: Here is Part 1 of this article.]

For more info on filtering spam in Outlook Express, visit Ara Rubyan's Spam Blocker Central.

Ara Rubyan is experienced in training design and delivery and has been developing web based training for more than 10 years. His background in training coupled with his IT knowledge help him to ensure the seamless development of product user training.