Prakash
my PhD thesis deals with the application of Mutual Information
to determine the strength of dependencies between random
variables in Bayesian networks. The thesis shows results of how
BN evaluation algorithms can be optimied through the use of MI.
So, if you're interested in applications, please look at
http://www.csse.monash.edu.au/~njitnah/Thesis/
Nathalie
> Judea Pearl wrote:
> >
> > Prakash,
> > Shannon's mutual information has some nice and some
> > peculiar properties, see pages 321-323, Probabilistic Reasoning
> > (1988)
> > ==========Judea
> >
> > To: [EMAIL PROTECTED]
> > Subject: [UAI] Measures of dependence between random variables
> > From: "Prakash P. Shenoy" <[EMAIL PROTECTED]>
> > Date: Wed, 24 May 2000 11:25:20 -0700
> > Sender: [EMAIL PROTECTED]
> > Precedence: bulk
> > Content-Type: text
> > Content-Length: 384
> >
> > Dear UAI colleagues,
> >
> > Could someone please point me to references on different measures of
> > dependence between random variables? I am familiar with
> > covariance/correlation coefficient that measures linear dependence between
> > variables. What else is out there that is being used by the Bayes net
> > learning community? Thanks in advance for the pointers.
> >
> > Prakash Shenoy
> > <[EMAIL PROTECTED]>
- --
Nathalie Jitnah, Research Fellow
School of Computer Science and Software Engineering
Monash University, Clayton, VIC 3800, Australia
email: [EMAIL PROTECTED], ph:+61 3 9905 5823, fax:+61 3 9905 5146
------- End of Forwarded Message