>>>>> "DP" == Daniel Pittman <[EMAIL PROTECTED]> writes:
>> Rule optimization is proceeding. You might find a better/faster regex >> engine, but you'll probably have to re-optimize the rules for that >> engine vs the perl engine. I think we're going to be focussing on In one of my previous lives, we used a system called LMDS from this large defense contractor called Logicon. This system was essentially designed from some intelligence agencies to scan large volumes of news feeds rapidly. Believe me, it was fast as anything I've ever seen. What you did was define a profile for each interest area. The profile was a bunch of keywords to search for with limits such as "within N words of" or "not when this other word appears", and the like. What the system then did was compile all of these profiles into one big custom search object which matched all profiles simultaneously. When you fed in a message, it returned a list of the names of every profile that it matched. On a dual Pentium II 400MHz box running BSD/OS, I could cram about 300 messages averaging 2 to 4 kbytes into it to match against about 500 profiles and have the results within about 2 minutes. It seems to me that such technology could be applied here, but I don't know of any freely availble software to do such a merged search. It would make for a great masters project... -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= 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/ _______________________________________________________________ Have big pipes? SourceForge.net is looking for download mirrors. We supply the hardware. You get the recognition. Email Us: [EMAIL PROTECTED] _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
