I get 98-98.5% accuracy with POPfile [sourceforge.net]. I get about 200 mails a day, of which around 30% spam. I get about 1 false negative a day, and maybe 2 or 3 false positives a month. It's a personal solution and as such is much more attractive to me than something server-based which has to be installed by a [typically VERY uncooperative] BOFH.
I use it experimentally for general mail classification (business/personal/a variety of mailing lists etc., all in all 7 buckets) on my home machine, and it works fine in these conditions too, although the accuracy is a bit lower (around 95%).
Is the 1.5% difference worth you paying for the bandwidth and CPU cycles it takes to identify those spam emails on your client, or would you rather make the spammer waste hours/days trying to get a single piece of email to you without costing you much more than a "Try again later" message?
The CPU cycles aren't a problem. I'm working on a 2+ GHz CPUs anyway (I wouldn't be able to do my normal job without them), and performing Bayesian classification for 3-4 seconds 10 times a day is not something I'd notice.
I can see however bandwidth possibly becoming a problem, if two years from now I start getting 10x more spam and work on a dialup...
"Well hello there Charlie Brown, you blockhead."
-- Lucy Van Pelt
97%? not impressive. It's POPfile for me (Score:4, Informative)
I use it experimentally for general mail classification (business/personal/a variety of mailing lists etc., all in all 7 buckets) on my home machine, and it works fine in these conditions too, although the accuracy is a bit lower (around 95%).
98.5% sure...but you're paying for it (Score:1)
Re:98.5% sure...but you're paying for it (Score:2)
I can see however bandwidth possibly becoming a problem, if two years from now I start getting 10x more spam and work on a dialup...