From quinlan@pathname.com Mon Sep 2 12:32:26 2002 Return-Path: Delivered-To: yyyy@localhost.netnoteinc.com Received: from localhost (localhost [127.0.0.1]) by phobos.labs.netnoteinc.com (Postfix) with ESMTP id 9001A43F99 for ; Mon, 2 Sep 2002 07:32:24 -0400 (EDT) Received: from phobos [127.0.0.1] by localhost with IMAP (fetchmail-5.9.0) for jm@localhost (single-drop); Mon, 02 Sep 2002 12:32:24 +0100 (IST) Received: from proton.pathname.com (adsl-216-103-211-240.dsl.snfc21.pacbell.net [216.103.211.240]) by dogma.slashnull.org (8.11.6/8.11.6) with ESMTP id g7V0KkZ16711 for ; Sat, 31 Aug 2002 01:20:47 +0100 Received: from quinlan by proton.pathname.com with local (Exim 3.35 #1 (Debian)) id 17kw0Z-0001k9-00; Fri, 30 Aug 2002 17:20:59 -0700 To: yyyy@spamassassin.taint.org (Justin Mason) Cc: SpamAssassin-devel@sourceforge.net Subject: Re: [SAdev] GA-assigned SPAM_PHRASE_55_XX score References: <20020830103221.9148E43F99@phobos.labs.netnoteinc.com> From: Daniel Quinlan Date: 30 Aug 2002 17:20:59 -0700 In-Reply-To: yyyy@spamassassin.taint.org's message of "Fri, 30 Aug 2002 11:32:16 +0100" Message-Id: X-Mailer: Gnus v5.7/Emacs 20.7 jm@jmason.org (Justin Mason) writes: > BTW I tried tweaking some of the scores that lint-rules complained about > being negative when they shouldn't be, and it *ruined* the results. it's > worth hand-tweaking a bit, but in some cases, there's counter-intuitive > combinatorial effects like the above, so be careful when tweaking; run > a "./logs-to-c && ./evolve -C" to check the new hitrates afterwards. My tendency is to say that we shouldn't tweak at all. - Dan