--On Friday, October 13, 2006 9:23 AM +0100 Justin Mason <[EMAIL PROTECTED]> wrote:

Please bear in mind, also, that there are 5 different rules that
use RFCI data, and they have wildly varying accuracies and scores:

SPAM%    HAM%    S/O    RANK    SCORE   NAME
 3.7247  0.0540  0.986  0.85    2.60    DNS_FROM_RFC_DSN                
 2.2447  0.1700  0.930  0.73    1.94    DNS_FROM_RFC_BOGUSMX            
15.1533  4.6068  0.767  0.51    1.45    DNS_FROM_RFC_POST               
18.6219  8.6003  0.684  0.49    1.71    DNS_FROM_RFC_ABUSE              
6.4258   4.0476  0.614  0.48    0.20    DNS_FROM_RFC_WHOIS

DNS_FROM_RFC_DSN fires on 3.7247% of spam, and only 0.054% of ham, giving
it an accuracy of 98.6%.

OTOH, DNS_FROM_RFC_POST, DNS_FROM_RFC_ABUSE, and DNS_FROM_RFC_WHOIS will
likely not make it into the next release going by those rates.

Rather than remove them, would it make sense to rescore them with a much lower weight, perhaps in some automated way? Even if the rules were useless, it might be desirable to give them a "report only" score (I think 0.001?) for the human who reviews the reports.

Cc'ing to the dev list since I'm raising the issue of changing the mass-check machinery.

Reply via email to