On Mon, 20 Oct 2003 18:45:03 -0700, "albinali" <[EMAIL PROTECTED]>
wrote:

> Hi all,
>   Thanks for the responses, I will try and explain the problem more clearly
> using a different example:
[ ... ]
>              .... The data set set collected from the space has
> sensor readings along with the activities. Moreover, I want to construct a
> bayesian network for every activity. 

Is this something I should know, 
'What's a Bayesian network'?
 
>                             Notice that if I include sensors that
> are somehow linearly dependant, the bayesian network will get multiple
> evidence from the same source, so thats why I would like to eliminate highly
> correlated variables. The approach that I am using to tackle this problem,
> is :

Or, I can put it another way.  You are implying you need
to meet some difficult and unreasonable conditions in 
order to create a "bayesian network."  

Why create that?  It is not enormously familiar to
statisticians, to judge by the few hits from groups.google 
in  the sci.stat.*   groups.  It is not familiar to me.  However,
I do note that there's a huge number of hits on the Web, and 
a  *lot*  more hits on  comp.ai.*    than sci.stat.*  -- that seems
to be a tool  of the a.i.  community, more than the statistical.
You might post there.

I was my impression, as I posted before, that you ought
to learn more about what you are trying to measure,
and figure out the  *dimensions*  and then some
interval scores;  so that regression  tools will be applicable.
Of course, regression tools are ones that I am familiar with.

Hope this helps.
-- 
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
"Taxes are the price we pay for civilization." 
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to