Hello Martin,

Sunday, January 18, 2004, 3:58:34 AM, you wrote:

>> L_MIME_BOUND_MANY_DIG -- 153s/0h of 92209 corpus (74874s/17335h) 01/17/04

MR> I've already modified the repeat counts for number of digits in
MR> L_MIME_BOUND_MANY_DIG, since it's not constant:

MR> header L_MIME_BOUND_MANY_DIG        Content-Type =~ /boundary=\"\d{15,}\"/

MR> changed from \d{19,} to \d{15,}

I've changed the copy in my files to 13.  Frequencies:

Section 3 -- Frequencies Log
(First numeric frequencies, followed by percentage frequencies)

OVERALL     SPAM      HAM     S/O   SCORE  NAME
  91714    74113    17601    0.808   0.00    0.00  (all messages)
    688      688        0    1.000   1.00   3.00  MR_MIME_BOUND_MANY_DIG14
    688      688        0    1.000   1.00   3.00  MR_MIME_BOUND_MANY_DIG13
    688      688        0    1.000   1.00   3.00  MR_MIME_BOUND_MANY_DIG15
    688      688        0    1.000   1.00   3.00  MR_MIME_BOUND_MANY_DIG11
    688      688        0    1.000   1.00   3.00  MR_MIME_BOUND_MANY_DIG12
    596      596        0    1.000   0.79   3.00  MR_MIME_BOUND_MANY_DIG16
    485      485        0    1.000   0.55   3.00  MR_MIME_BOUND_MANY_DIG17
    354      354        0    1.000   0.25   3.00  MR_MIME_BOUND_MANY_DIG18
    241      241        0    1.000   0.00   3.00  MR_MIME_BOUND_MANY_DIG19

OVERALL%   SPAM%     HAM%     S/O    RANK   SCORE  NAME
  91714    74113    17601    0.808   0.00    0.00  (all messages)
100.000  80.8088  19.1912    0.808   0.00    0.00  (all messages as %)
  0.750   0.9283   0.0000    1.000   1.00    3.00  MR_MIME_BOUND_MANY_DIG14
  0.750   0.9283   0.0000    1.000   1.00    3.00  MR_MIME_BOUND_MANY_DIG13
  0.750   0.9283   0.0000    1.000   1.00    3.00  MR_MIME_BOUND_MANY_DIG15
  0.750   0.9283   0.0000    1.000   1.00    3.00  MR_MIME_BOUND_MANY_DIG11
  0.750   0.9283   0.0000    1.000   1.00    3.00  MR_MIME_BOUND_MANY_DIG12
  0.650   0.8042   0.0000    1.000   0.79    3.00  MR_MIME_BOUND_MANY_DIG16
  0.529   0.6544   0.0000    1.000   0.55    3.00  MR_MIME_BOUND_MANY_DIG17
  0.386   0.4776   0.0000    1.000   0.25    3.00  MR_MIME_BOUND_MANY_DIG18
  0.263   0.3252   0.0000    1.000   0.00    3.00  MR_MIME_BOUND_MANY_DIG19

I guess I could have gone down as low as 11 (no ham, same spam), but I
figured leaving a few more digits would help avoid future ham.

Bob Menschel





-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk

Reply via email to