Hello MikeD, Sunday, November 16, 2003, 6:28:41 PM, you wrote, among other things:
M3> First, you are on the current version of Bayesit, 0.4gm, right? Yes M3> Anyway, once I got it filtering, I have been steadily using the M3> "special/mark as junk" to classify those messages that it missed (I have M3> yet to find anything in the junk folder that did not belong ... no false M3> positives, a good thing) And now I am slowly lowering the threshold to M3> see how far I can go before I start to get false positives (I am M3> currently down to 60 <g>) I have now done some testing by sending Spam messages to myself and they do get filtered. The 'selective filtering' is however another story. M3> The one thing I would like to be able to see is what score certain M3> messages did get so that I could have a better idea how well it is M3> classifying messages <sigh> Agreed. The macro for measuring and displaying the spaminess of a word so far has only produced a zero result, even on words like Viagra! -- Best regards, Roger Using The Bat! v2.01.26 on Windows ME, v4.90 Build 3000 BayesIt! 0.4gm ________________________________________________________ Current beta is 2.01.26 | "Using TBBETA" information: http://www.silverstones.com/thebat/TBUDLInfo.html
