-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello Justin,
SB> At 16:58 17/07/03 -0700, Justin Mason wrote: >>> My hunch is that auto-learning waters down the effectiveness of >>> manual training. ... >>However, I'd say at that kind of volume, there really isn't much need >>to keep auto-learning if you don't want to; you could also turn it off >>and just train on what's reported as FPs/FNs and that should work OK. SB> What I was starting to think about, is maybe there could be a SB> different weighting for auto-learning as compared to manual learning? Wouldn't this be easily accomplished by just increasing the auto-learn safety margin? Raise the score at which spam is auto-learned, and lower the score at which ham is auto-learned, and you'll auto-learn fewer messages. If I understand how this works, the manual teaching will therefore rate higher since it will have less competition within the database. Bob -----BEGIN PGP SIGNATURE----- Version: PGP 8.0 iQA/AwUBPxdNLJebK8E4qh1HEQLx5ACeJe6KUOAI+AxL8s3Mt1MIxTIpxLsAnjbT MOto9K6zI2nCbe5s8rQ5Gh20 =H4ku -----END PGP SIGNATURE----- ------------------------------------------------------- This SF.net email is sponsored by: VM Ware With VMware you can run multiple operating systems on a single machine. WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines at the same time. Free trial click here: http://www.vmware.com/wl/offer/345/0 _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
