It was probably me when I broght all the 2_3_0 line changes forward onto the
trunk.
C
Daniel Quinlan wrote:
DQ> Bart Schaefer <[EMAIL PROTECTED]> writes:
DQ>
DQ> > Hmm, I just did a "cvs up" (on the head, not the branch) and:
DQ> >
DQ> > score: FORGED_RCVD_TRAIL 1.000 -> absent
DQ> > score: MSG
On Saturday 15 June 2002 12:31 pm, Craig R Hughes wrote:
> Michael Moncur wrote:
> MM> score ASCII_FORM_ENTRY -1.660
>
> Looks like lots of false positives on the appended lines at the bottom of
> Sourceforge mailing list messages. This score should probably be pumped up
> a little
On Sat, Jun 15, 2002 at 12:31:31PM -0700, Craig R Hughes wrote:
> I think the rule needs to be adjusted to not trigger on 3 words' presence in the
> message, since "asian" and "hardcore" can occur in legitimate messages.
> Instead, it should trigger based on %age of words which are in the list, so
On Sat, 2002-06-15 at 14:31, Craig R Hughes wrote:
> MM> - Not as weird as all that, apparently
> MM> score MSGID_CHARS_WEIRD -2.178
>
> Looks like mail servers (Exchange and Netscape mail server) sometimes
> create
> message ids which look like:
>
> Message-Id:
>
> I don't know w
Michael Moncur wrote:
MM> When a new release comes out I like to be anal-retentive and go through the
MM> GA second-guessing its scores. This is my report for 2.30.
A valuable service we've come to count on.
MM> - RATWARE must be fixed, it was negative last time
MM> score RATWARE
Bart Schaefer <[EMAIL PROTECTED]> writes:
> Hmm, I just did a "cvs up" (on the head, not the branch) and:
>
> score: FORGED_RCVD_TRAIL 1.000 -> absent
> score: MSGID_CHARS_WEIRD -2.178 -> absent
> score: FROM_ADDRESS_EQ_REAL 1.000 -> absent
> score: X_NOT_PRESENT -1.920 -> absent
> score: FROM_A
Craig R Hughes writes:
> So it's just because the GA could get away with setting it to 0.921
> -- in practice it's a clear sign of nonspam, and we should just fix
> it at -2.0, which I've done on both branches now.
Okay. In HEAD, I made the rule less apt to be abused which is just as
well since
Daniel Quinlan wrote:
DQ> >>> score X_NOT_PRESENT -1.920
DQ> >>
DQ> >> This one is on my hitlist as well. Didn't work out very well.
DQ>
DQ> Craig R Hughes <[EMAIL PROTECTED]> writes:
DQ>
DQ> > But it actually turns out to be great at clawing back false
DQ> > positives. I think
Daniel Quinlan wrote:
DQ> Craig R Hughes <[EMAIL PROTECTED]> writes:
DQ>
DQ> >> score: BUGZILLA_BUG -2.000 -> 0.921
DQ>
DQ> > Moved to the right section of the scores file, and score reverted to -2.0
DQ>
DQ> But why is it positive? Doesn't it mean there are good messages in
DQ> the spam corpus o
>>> score X_NOT_PRESENT -1.920
>>
>> This one is on my hitlist as well. Didn't work out very well.
Craig R Hughes <[EMAIL PROTECTED]> writes:
> But it actually turns out to be great at clawing back false
> positives. I think we should leave it in with the low score.
Hrmm, it
On Sat, 15 Jun 2002, Craig R Hughes wrote:
> Bart Schaefer wrote:
>
> BS> These look suspicious:
>
> Changed back to ...
Hmm, I just did a "cvs up" (on the head, not the branch) and:
score: FORGED_RCVD_TRAIL 1.000 -> absent
score: MSGID_CHARS_WEIRD -2.178 -> absent
score: FROM_ADDRESS_EQ_REAL
Bart Schaefer wrote:
>> These look suspicious:
>>
>> score: ASCII_FORM_ENTRY 0.036 -> -1.660
Craig R Hughes <[EMAIL PROTECTED]> writes:
> Changed back to 0.5 -- as mentioned in previous message, this is
> triggering on the sourceforge-appended footers on mailing list
> mails.
Maybe it would b
Daniel Quinlan wrote:
DQ> Michael Moncur <[EMAIL PROTECTED]> writes:
DQ>
DQ> > And a few slightly questionable scores:
DQ> >
DQ> > - This was 0.87 before. Less and less useful?
DQ> > score FROM_AND_TO_SAME -2.071
DQ>
DQ> I think this one should go. It's a common way to send email t
Bart Schaefer wrote:
BS> These look suspicious:
BS>
BS> score: ASCII_FORM_ENTRY 0.036 -> -1.660
Changed back to 0.5 -- as mentioned in previous message, this is triggering on
the sourceforge-appended footers on mailing list mails.
BS> score: BUGZILLA_BUG -2.000 -> 0.921
Moved to the right sect
Bart Schaefer <[EMAIL PROTECTED]> writes:
> TO_ADDRESS_EQ_REAL happens a _lot_ with Outlook and Outlook Express. Any
> time an OE user receives a message with no real name, the address part
> gets added to their address book as the name. If they later send a reply
> or other message to that add
On 15 Jun 2002, Daniel Quinlan wrote:
> Bart Schaefer <[EMAIL PROTECTED]> writes:
>
> > score: FORGED_RCVD_TRAIL absent -> 1.000
> > score: FROM_ADDRESS_EQ_REAL absent -> 1.000
> > score: TO_ADDRESS_EQ_REAL absent -> 1.000
>
> You're looking at HEAD. These are new rules I added last night.
Ah
Bart Schaefer <[EMAIL PROTECTED]> writes:
> score: ASCII_FORM_ENTRY 0.036 -> -1.660
> score: BUGZILLA_BUG -2.000 -> 0.921
BUGZILLA_BUG obviously needs to be fixed. Maybe an eval would be best.
> score: DATE_MISSING 0.248 -> -2.140
> score: EXCUSE_16 1.345 -> -0.721
> score: FORGED_HOTMAIL_RCVD
Michael Moncur <[EMAIL PROTECTED]> writes:
> And a few slightly questionable scores:
>
> - This was 0.87 before. Less and less useful?
> score FROM_AND_TO_SAME -2.071
I think this one should go. It's a common way to send email to a
large list of people without subjecting them all
On Sat, 15 Jun 2002, Michael Moncur wrote:
> When a new release comes out I like to be anal-retentive and go through
> the GA second-guessing its scores. This is my report for 2.30.
In a similar vein, here are the significant score changes since the last
CVS version before the GA was re-run. (I
On Sat, 15 Jun 2002 the voices made Michael Moncur write:
> When a new release comes out I like to be anal-retentive and go through the
> GA second-guessing its scores. This is my report for 2.30.
> - This works well for me but users in some countries may want to change it
> score SUBJ_FULL_OF_8
When a new release comes out I like to be anal-retentive and go through the
GA second-guessing its scores. This is my report for 2.30.
Overall, the GA did a NICE job this time. I have very little to complain
about and haven't found a single score I'll be bothering to override. Here
are a few scor
21 matches
Mail list logo
