-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Daniel Quinlan writes:
>RHSBL and overlap:
>
>17934   0.871   0.092   T_RCVD_IN_AHBL_RHSBL,__RCVD_IN_SBL_XBL
>14374   0.698   0.090   T_RCVD_IN_AHBL_RHSBL,__RCVD_IN_SORBS
>13134   0.638   0.074   T_RCVD_IN_AHBL_RHSBL,RCVD_IN_XBL
>10697   0.520   0.071   T_RCVD_IN_AHBL_RHSBL,RCVD_IN_DSBL
>10612   0.516   0.096   T_RCVD_IN_AHBL_RHSBL,RCVD_IN_BL_SPAMCOP_NET
>7454    0.362   0.079   T_RCVD_IN_AHBL_RHSBL,RCVD_IN_SORBS_DUL
>
>Lower overlap than the others, this might be worth keeping (and it's a
>separate query anyway).

I would say keep RHSBL, and leave the rest.  In particular, the blocking
of an entire Spanish ISP is a big "thumbs down" factor for me.  I'm
even a little lukewarm on RHSBL, too, just because it's another DNS
query and 9.11% for a DNSBL is not exactly stellar.

>Maybe we should let the perceptron take a whack at it.  So, why aren't
>we running the perceptron on nightly/weekly results?  ;-)

go for it!  bugzilla.spamassassin.org is crying out for new uses for
CPU time ;)

- --j.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
Comment: Exmh CVS

iD8DBQFAjYiQQTcbUG5Y7woRAnp2AKCPQs7v0QH4f1UrEqrkYgwmdhY56gCg3KKH
wSFNLZGFrDxunqsuMhfHuD0=
=xY3j
-----END PGP SIGNATURE-----

Reply via email to