>>>>> "KGM" == Kingsley G Morse, <Kingsley> writes:
KGM> How about sampling the network checks, so that instead KGM> of 400,000, only doing, say 500? KGM> It seems to me that sampling a few hundred network KGM> checks would arrive at a better score for them than KGM> hand coding. The problem I see with evolving the scores for network tests is that the BL's are temporal -- that is, the hit/miss status of any given message in the corpus can change at any time as the database changes. Sites are added and removed from the various BL's all the time, and so you can't get consistent and reliable results, especially with the older spam in the corpus whose relay-raped servers are since closed. As for how long a message stays in razor, I don't know, so perhaps that one could be evolved. It's a tough call since there are so many false-positives in Razor. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Vivek Khera, Ph.D. Khera Communications, Inc. Internet: [EMAIL PROTECTED] Rockville, MD +1-240-453-8497 AIM: vivekkhera Y!: vivek_khera http://www.khera.org/~vivek/ _______________________________________________________________ Don't miss the 2002 Sprint PCS Application Developer's Conference August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
