| About Spam Assassin™ For Email Spam Filtering |
| |
| FAQ |
 |
Q: How does this affect me?
A: In short it flags messages it thinks are spam. The amount of spam you have to wade through will be reduced significantly if you filter those messages. All email messages received by our servers will be scanned for SPAM signatures. Email flagged as SPAM will have the text *****Spam Alert***** inserted into the subject line of the message. |
 |
Q: What do I do with email that is marked with *****Spam Alert***** ?
A: You can do several things with it: delete it, filter it to a special folder, or do nothing. |
 |
Q: I use WebMail. How do I filter *****Spam Alert***** to a special folder?
A: Sorry currently Webmail do not have built in filters. |
 |
Q: How do I filter messages with Outlook, Outlook Express, Netscape Mail, or Eudora?
A: The steps for creating filters can be found in the help documentation for each of these programs, and are easy setting up. If you need further assistance, let us know. |
 |
Q: How effective is this new AntiSPAM software?
A: In our real-time, real-world tests, it kills 80% of SPAM on contact. Several domains receiving hundreds of SPAM messages per day were used to test the software, and we feel this is the best technology to come out yet, primarily because of the high SPAM flagging rate versus the false positive rate (a.k.a. legitimate messages getting marked as SPAM) |
 |
Q: What do the numbers after *****Spam Alert***** mean?
A: The first number is the scoring or rating given to a message, with the second number being the threshold at which the message will be flagged as SPAM. The threshold is set to 6.0, which in our tests, catches greater than 70% of all SPAM. |
 |
Q: Does the threshold ever change?
A: The threshold does not change, but at some point in the near future, you will be able to set the threshold for individual mailboxes with an upgraded hosting account options. |
 |
Q: What determines how many "points" an email gets?
A: The message analyzer engine is based on a new technology employing Bayesian Statistics, which first appeared in a piece of software called Spam Assassin. Bayesian Statistics use both rules and dynamic learning to increase the chance of detecting SPAM. You will notice that messages marked as SPAM have new email headers that show what rules were triggered and for how many points. |
 |
Q: This new AntiSPAM engine learns? As in Artificial Intelligence?
A: Well, yes and no. It still depends on us for determining WHAT to learn from (see below) but once it has that, yes, it will learn on it's own and become more effective with time. |
 |
Q: How can I help?
A: There are two distinct things you can do. The first is to send any emails that are marked as SPAM that should not have been to us. The second is to send us messages that slipped through without getting marked as SPAM that should have been. If you still have a high occurrence of unmarked SPAM (more than 30 a day), please contact us, as there are additional things we can on a per account basis. |