It just goes to show you that all things come back around.  I had a pretty
good group looking at just these issues at USC about 10-15 years ago.  At
that point, our biggest challenge was convincing the UAI community that the
issue was of more than tangential interest (which was a lot further than we
were able to get with the general AI community which was then in the last
stages of its "if it's got numbers, it's not AI" phase).  Anyway, both
Keung-Chi Ng and Izhar Matzkevich wrote their dissertations on topics
relating to the combination of Bayes nets.  KC's approach was primarily
experimental and Izhar's was essentially theoretical.  I'm sure that someone
at USC ought to be able to dust off a copy for you.  In the meantime, you
might want to try the following articles:

        1.      Izhar Matzkevich and Bruce Abramson "Decision-Analytic
Networks in Artificial Intelligence." Management Science 41(1), 1-22, 1995. 
        2.      Keung-Chi Ng and Bruce Abramson "Probabilistic
Multi-Knowledge-Base Systems." Journal of Applied Intelligence 4(2), 19-236,
1994.
        3.      Izhar Matzkevich and Bruce Abramson. "Some Complexity
Considerations in the Combination of Belief Networks." In Uncertainty in
Artificial Intelligence - 93, D. Heckerman and A.J. Mamdani, eds. (San
Mateo, CA: Morgan Kaufmann, 1993), 152-158.
        4.      Izhar Matzkevich and Bruce Abramson. "Deriving a Minimal
I-Map of a Belief Network Relative to a Target Ordering." In Uncertainty in
Artificial Intelligence - 93, D. Heckerman and A.J. Mamdani, eds. (San
Mateo, CA: Morgan Kaufmann, 1993), 159-165. 
        5.      Bruce Abramson and Keung-Chi Ng.  "Consensus Diagnosis: A
Simulation Study."  IEEE Transactions on Systems, Man, and Cybernetics
22(5), 916-928, 1992.
        6.      Izhar Matzkevich and Bruce Abramson.  "The Topological
Fusion of Bayes Nets." In Uncertainty in Artificial Intelligence - 92, D.
Dubois, M.P. Wellman, B. D'Ambrosio, and P. Smets, eds. (San Mateo, CA:
Morgan Kaufmann, 1992), 191-199.
Bruce Abramson
Charles River Associates
1201 F St. NW, Suite 700
Washington, DC 20004-1204
Phone: 202-662-3800
Direct: 202-662-3871
Fax: 202-662-3910
www.crai.com


On Wed, 18 Sep 2002, Sergio Storari wrote: 
Hi I'm beginning to work on multi-agent system for distributed diagnosis. To
this purpose I'm investigating on bayesian networks and how to merge the
reasoning of different agents each one enclosing its own bayesian network
and collecting documentation about works regarding these methodologies. I
kindly ask to anyone who knows relevant literature about this subject to
send me references. Thank you in advance. 
Best Regards 
Sergio Storari 
Ing.Sergio Storari 
Dipartimento di Ingegneria 
 Universit� di Ferrara 
 Via Saragat,1 
 44100 Ferrara FE 
 Tel: +39532293893
 Fax: +39532768602  
email: [EMAIL PROTECTED]
</cgi-bin/compose?curmbox=F000000001&a=5725373f496042e98866f9830e3f0a68&mail
to=1&[EMAIL PROTECTED]&msg=MSG1032459880.190&start=1393857&len=4113&s
rc=&type=x> > ________________________________ 



Reply via email to