Spamishness

[Tim Diebert]

Greetings and Top of the Season to you all.

Because of my web based business, I have an email address available to all of nasty them spam spiders and such.

Because I need to not miss any business related emails, I have always refrained from using any spam filters. These filters are not 100% accurate...apparently, some 'good' emails can get nuked along with the bad ones. To prevent this from happening, I 'filter' my own mail and delete all the not wanted stuff. BUT...now what is happening is because I commonly get 400 emails a day, I am accidentally deleting good emails because of my over used and twitchy delete finger.

If I go away for a few days, it is not uncommon to have to sift through a thousand emails. It is now way over the top and I have to do something.
I am looking at using some form of filter.
Has anybody had a good experience with some software somewhere?
Thanks in advance for any help folks.

[megman]

Try this: Mailwasher (http://www.mailwasher.net/)

[Meerkat]

Yes! I use a free program that's integrated into Outlook. It is 100% accurate for "good" mail(1) and 90%+ accurate for spam. The 10% it's not sure about goes into a "maybe" folder for user inspection.

The program is called "SpamBayes" and it can be found at http://spambayes.sourceforge.net/ and it is actively developed. It not only runs on Outlook, there are versions for Linux/Unix and Mac too (it's not as simple to use on those two though).

Note 1: It will put "bad" mail in the inbox once in awhile if it has not encountered that kind of mail before, but once that mail is "delete as spam"-ed, it will do better. However, the spammers are always finding ways to change the "signature" of bad mail, so it's an ongoing struggle. I have never had a single piece of good mail get auto-trashed though.

[Tim Diebert]

Thanks guys.
I had checked out the Spam Washer yesterday and I will have a look at SpamBayes today.
Sounds like the ticket.
I always prefer to use even free software that other folks (I know) have tried and liked.
Appreciated.

[Meerkat]

Tim; Feel free to ask me for any assistance you might need in installing the program and it's supporting programs. It's not hard, but there are a couple of potential stumbling blocks. In particular, Python must be installed, and there's one step that the installer defaults incorrectly in many people's opinion (although it is the Microsoft standard default).

[Tim Diebert]

Thanks Meer. Soon as I get he time to install and run this thing I will let you know how it goes.

[TimH]

Any Bayesian filter should work pretty well once you "teach" it good from bad mail. Thats how Bayesian filters work. They learn and adapt to the new tricks of spammers.

Bayes' Rule provides a way of combining probabilities. The goal is to take two individual probabilities and combine them into a single value. If a message contains the word "sex," what is the likelihood the message is spam? If a message contains the word "male," what is its likelihood of being spam? And if the same message contains both "sex" and "male," then what is the probability the message is spam? The rule creates a mathematical framework for combining these individual values into a single probability for the e-mail message as a whole.

Technically, this scenario illustrates "naïve" Bayesian probability, where one assumes that the occurrence of one event is independent of the occurrence of the other, which generally isn't the case with spam, where, for example, "low" and "mortgage" occur together more frequently in spam than they do normally. Assuming nonindependence yields even better results, but it requires more complex computations.

Essentially, the Bayesian idea is to define the probability of one thing's happening if another thing is the case, in terms of the probability of the one, the probability of the other, and the reverse probability of the two. For example, if we want to know the probability that someone who smokes will get cancer, we need to know the number of cancer victims that are in the general population, the number of smokers, and the percentage of smokers among those who have cancer.

[Meerkat]

And if you skip all the technical details you don't really need to know to use SpamBayes, it boils down to the filter producing one of three values for each filtered email: "spam", "maybe spam" or (to use the 'in' term for good email) "ham". tongue.gif