Hello Marten, Thursday, May 18, 2006, 1:58:32 PM, you wrote:
MG> I have spent two days now feeding BayesIT and adjsuting the limits etc. MG> I have imported a whitelist and added to it, I have the regex filters MG> installed, I have imported a blacklist. MG> Still it fails to pick up half of the spams arriving. MG> Are tohers more successful with theior installations? I used to use it. I moved to Agvara ant think it (mostly) works better. I found that it took a **LOT** of mail to train it. I kept at least 1,000 spam and 1,000 ham messages around just to train it. I got it to better than 80% without using a white or black list and I *NEVER* got a false positive with it. Agvara gets about 95% of my spam but I do have to go through the spam because I get the occasional false positive with it. So the end result is that it works, but is a more 'finiky'. But it is also free <g> -- Best regards, MikeD mailto:[EMAIL PROTECTED] Using The Bat! v3.80.06 on Windows XP 5.1 Build 2600 Service Pack 2 ________________________________________________ Current version is 3.80.06 | 'Using TBUDL' information: http://www.silverstones.com/thebat/TBUDLInfo.html