[Spambayes] filter misclassification

Fri Apr 29 10:31:58 CEST 2005

Hello,

I obtained the following clues to understand why the filter misclassifies
mails from a particular source. Despite training such messages as "ham",
the Ham score is shown as "0". Kindly let me know how to force such a
message to be treated as "Ham".

Thanks in advance.

-gowranga

**************************************************************************
Combined Score: 50% (0.5)
**************************************************************************
Internal ham score (*H*): 0
Internal spam score (*S*): 0

# ham trained on: 0
# spam trained on: 0

**************************************************************************
0 Significant Tokens
**************************************************************************
token                               spamprob         #ham  #spam

**************************************************************************
Message Stream
**************************************************************************

Return-Path: <bounce-14810-111417 at lyris.msfc.nasa.gov>
X-Original-To: gowranga at serc.iisc.ernet.in
Delivered-To: gowranga at serc.iisc.ernet.in
Received: from iisc.ernet.in (iisc.ernet.in [144.16.64.3])
	by serc.iisc.ernet.in (Postfix) with ESMTP id 26B7F180F
	for <gowranga at serc.iisc.ernet.in>; Fri, 29 Apr 2005 09:54:31 +0530 (IST)
Received: from lyris.msfc.nasa.gov (www.spaceweather2.com [72.3.135.213])
	by iisc.ernet.in (8.12.9/8.12.8) with SMTP id j3TA0EBM048312
	for <gowranga at serc.iisc.ernet.in>; Fri, 29 Apr 2005 10:00:24 GMT
From: NASA Science News <snglist at lyris.msfc.nasa.gov>
To: NASA Science News <snglist at lyris.msfc.nasa.gov>
Subject: Prospecting for Lunar Water
Date: Thu, 28 Apr 2005 21:33:17 -0500
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="MIMEBoundarydf0496b01443e4ff2e2a4222f3c6cd6f"
List-Unsubscribe: <mailto:leave-snglist-111417C at lyris.msfc.nasa.gov>
Message-Id: <LYRIS-111417-14810-2005.04.28-21.33.19--gowranga#serc.iisc.ernet.in at lyris.msfc.nasa.gov>
X-Folder: Bulk
Status:
X-Status:
X-Keywords:

This is a multi-part message in MIME format.

--MIMEBoundarydf0496b01443e4ff2e2a4222f3c6cd6f
Content-Type: text/plain;
	charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit

NASA Science News for April 28, 2005

Settling alien worlds is thirsty work. Before sending people back to the Moon, NASA plans to send a robotic spacecraft first to hunt for water.

FULL STORY at

http://science.nasa.gov/headlines/y2005/28apr_lro.htm?list111417

The Science at NASA Podcast feed is available at http://science.nasa.gov/podcast.xml.

You are currently subscribed to snglist as: gowranga at serc.iisc.ernet.in.

This is a free service.

To unsubscribe click here: http://lyris.msfc.nasa.gov/u?id=111417C&n=T&l=snglist
or send a blank email to leave-snglist-111417C at lyris.msfc.nasa.gov

--MIMEBoundarydf0496b01443e4ff2e2a4222f3c6cd6f
Content-Type: text/html;
	charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit

<HTML><BODY>
NASA Science News for April 28, 2005
Settling alien worlds is thirsty work. Before sending people back to the Moon, NASA plans to send a robotic spacecraft first to hunt for water. 
FULL STORY at
<a href="http://science.nasa.gov/headlines/y2005/28apr_lro.htm?list111417">http://science.nasa.gov/headlines/y2005/28apr_lro.htm?list111417</a>
The Science at NASA Podcast feed is available at <a href="http://science.nasa.gov/podcast.xml.">http://science.nasa.gov/podcast.xml.</a> 
 
You are currently subscribed to snglist as: <a href="mailto:gowranga at serc.iisc.ernet.in">gowranga at serc.iisc.ernet.in</a>. 
This is a free service.
To unsubscribe click here: http://lyris.msfc.nasa.gov/u?id=111417C&n=T&l=snglist 
or send a blank email to <a href="mailto:leave-snglist-111417C at lyris.msfc.nasa.gov">leave-snglist-111417C at lyris.msfc.nasa.gov</a>
</body></HTML>

--MIMEBoundarydf0496b01443e4ff2e2a4222f3c6cd6f--

**************************************************************************
All Message Tokens
**************************************************************************
97 unique tokens'2005'
'28,'
'alien'
'april'
'are'
'as:'
'available'
'back'
'before'
'blank'
'cc:none'
'charset:iso-8859-1'
'click'
'content-type:multipart/alternative'
'content-type:text/html'
'content-type:text/plain'
'currently'
'email'
'email addr:serc.iisc.ernet.in.'
'email name:gowranga'
'feed'
'first'
'for'
'free'
'from:addr:lyris.msfc.nasa.gov'
'from:addr:snglist'
'from:name:nasa science news'
'full'
'header:Date:1'
'header:From:1'
'header:MIME-Version:1'
'header:Message-Id:1'
'header:Received:2'
'header:Return-Path:1'
'header:Subject:1'
'header:To:1'
'here:'
'hunt'
'message-id:@lyris.msfc.nasa.gov'
'moon,'
'nasa'
'news'
'people'
'plans'
'podcast'
'proto:http'
'reply-to:none'
'robotic'
'science'
'science at nasa'
'send'
'sender:none'
'sending'
'service.'
'settling'
'skip:l 40'
'snglist'
'spacecraft'
'story'
'subject: '
'subject:Lunar'
'subject:Prospecting'
'subject:Water'
'subject:for'
'subscribed'
'the'
'thirsty'
'this'
'to:2**0'
'to:addr:lyris.msfc.nasa.gov'
'to:addr:snglist'
'to:name:nasa science news'
'unsubscribe'
'url:111417c'
'url:28apr_lro'
'url:gov'
'url:headlines'
'url:htm'
'url:id'
'url:l'
'url:list111417'
'url:lyris'
'url:msfc'
'url:n'
'url:nasa'
'url:podcast'
'url:science'
'url:snglist'
'url:t'
'url:u'
'url:xml'
'url:y2005'
'water.'
'work.'
'worlds'
'x-mailer:none'
'you'

On Fri, 1 Apr 2005, Tony Meyer wrote:

> > However, mails from "NASA" keep getting delivered to "spam"
> > folder, despite the fact I train/retrain the filter not to
> > treat it as "spam".
>
>
> You can see the clues list via the web interface (you can use sb_upload or
> just sb_server and use the form on the main page), or there's a script to
> print them out in CVS (and will be in 1.1) in the contrib/ directory, or if
> you turn on the X-Spambayes-Evidence header they will be there.
>
> =Tony.Meyer