• To: 'Bill de hOra' <bill.dehora@p...>, XML Developers List <xml-dev@l...>
• Subject: RE: Something altogether different?
• From: "Bullard, Claude L (Len)" <len.bullard@i...>
• Date: Mon, 25 Apr 2005 09:18:57 -0500

Thanks Bill.  Also particle and wiener filters (right back to Bayes).

Rocket science.  If the predict/update rate is fast enough given the 
distance, you are right.   I wonder about the performance as the 
number of terms increases but that's why we have Moore's Law. 

The interesting bit to me is that given multiple 
notations (even if all XML), the same similarity 
metric should be returned and if persistent, can 
be identified by a URI.  Not new news.

len


From: Bill de hOra [mailto:bill.dehora@p...]

Bullard, Claude L (Len) wrote:

> Take the vector measures and tie them together with 
> URIs across multiple notations for the same observations 
> and that is an interesting system for machine learning 
> as has been shown time and time again.  They aren't 
> as useful for targeting munitions;

You might want to look up Kalman Filters.

• Prev by Date: Re: Something altogether different?
• Next by Date: RE: Something altogether different?
• Previous by thread: RE: Something altogether different?
• Next by thread: RE: Something altogether different?
• Index(es):
  • Date
  • Thread