Critical Eye on SpamAssassin

ErrorBase writes "In this Infoworld article, Logan G. Harbaugh makes a great deal about an ancient (2.44) version of SpamAssassin comparing it with newer comercial variants. Quote : You get what you pay for. [...] However, it took more than 10 times as long to install and configure SpamAssassin as it did any of the other products. " Why did he not ask Kevin Railsback who had the whole thing working some while ago?)"

Comments

The algorithm

[Mr_Silver]

I know that this will probably be modded off topic as it is a little, but I want to adapt the bayes algorithm to recommend television programs from a schedule.

The idea being that good programmes are classed as, if you like, spam and bad ones aren't. Therefore anything that gets tagged can be considered a programme that might be interesting to the user.

The problem is that I've not seen a good *basic* description of how the algorithm works (i'll be implementing it in Perl with a view to porting it to other languages). Preferably with some sample values and a step by step guide on how the final score comes out.

Can anyone point me to a resource? Paul Grahams description is good, but the formula makes no sense and there aren't any examples.