POPFile after two months

I’ve been using POPFile on my Linux system for right at two months now.  Its statistics show that it has classified 2,724 messages in that time, with an accuracy of 96.91%.  Of the 84 classification errors, 32 of them have been false positives:  good messages that POPFile identified as spam.  Most of those errors occurred in the first few weeks that I was using POPFile while I trained it to tell the difference between ham and spam.  I still scan POPFile’s classification, but it’s been over a week since I last got a false positive.  I’m going to reset the statistics on August 1 and see how it fares over the next couple of months.

I’ve found that it’s a whole lot easier to scan POPFile’s classification results for false positives than it is to try sorting out my inbox by hand.  If you’re having spam trouble (and who isn’t these days?), I’d highly recommend giving POPFile a look.