You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@spamassassin.apache.org by Jari Fredriksson <ja...@iki.fi> on 2012/08/08 23:31:16 UTC
masscheck corpus quality
How do you read this? This is from ruleqa page for today's corpus.
I have lots more ham messages than 3119. And it shows no ham for this
year. What? I have 11230 ham messages in my corpus...
jarif Spam messages Score range Ham messages Score
range
in 2005 0 80 (0%)
[0,2]
in 2006 0 9 (0%)
[0,1]
in 2007 0 21 (0%)
[0,2]
in 2008 0 12 (0%)
[0,0]
in 2009 0 73 (0%)
[0,5]
in 2010 0 474 (0%)
[0,5]
in 2011 2 (0%) [2,8] 2450 (1%)
[0,8]
in 2011-09 1 (0%) [3,3]
0
in 2011-12 3 (0%) [0,2]
0
in 2012-01 16 (0%) [0,12]
0
in 2012-02 13 (0%) [0,46]
0
in 2012-03 20 (0%) [0,28]
0
in 2012-04 75 (0%) [0,66]
0
in 2012-05 295 (0%) [0,49]
0
TOTAL: 425 (0%) [0,66] 3119 (2%)
[0,8]
--
"Unfortunately suspend does mean things sometimes"
Husse Apr 25 2007
Re: masscheck corpus quality
Posted by Jari Fredriksson <ja...@iki.fi>.
09.08.2012 04:20, John Hardin kirjoitti:
> On Thu, 9 Aug 2012, Jari Fredriksson wrote:
>
>> How do you read this? This is from ruleqa page for today's corpus.
>>
>> I have lots more ham messages than 3119. And it shows no ham for this
>> year. What? I have 11230 ham messages in my corpus...
>
> What numbers do you see when you hover over the hit % for your corpora
> in the "set 0, broken down by contributor" section? Are they similarly
> underreported?
>
Yes. But let's see how it looks when current day stats show up. I think
I have had local problems causing all of this.
--
You have an unusual magnetic personality. Don't walk too close to
metal objects which are not fastened down.
Re: masscheck corpus quality
Posted by John Hardin <jh...@impsec.org>.
On Thu, 9 Aug 2012, Jari Fredriksson wrote:
> How do you read this? This is from ruleqa page for today's corpus.
>
> I have lots more ham messages than 3119. And it shows no ham for this
> year. What? I have 11230 ham messages in my corpus...
What numbers do you see when you hover over the hit % for your corpora in
the "set 0, broken down by contributor" section? Are they similarly
underreported?
--
John Hardin KA7OHZ http://www.impsec.org/~jhardin/
jhardin@impsec.org FALaholic #11174 pgpk -a jhardin@impsec.org
key: 0xB8732E79 -- 2D8C 34F4 6411 F507 136C AF76 D822 E6E6 B873 2E79
-----------------------------------------------------------------------
No representation without taxation!
-----------------------------------------------------------------------
7 days until the 67th anniversary of the end of World War II