Dear members of the UAI community, We are currently writing a paper about using repeated patterns (which we call "idioms") to develop large scale Bayesian networks in a more efficient way. We try to identify the basic building blocks that will be sufficient to construct any Bayesian network. These can then be instantiated in many different contexts with different meaning. Does anybody know if such issues have been investigated before? The paper will also look into using OOBNs (object-oriented Bayesian networks) to support this framework. Greetings, Lars -- Lars M. Nielsen HUGIN Expert A/S Email: [EMAIL PROTECTED] Niels Jernes Vej 10 Phone (o): +45 96 35 45 48 DK-9220 Aalborg East Phone (h): +45 98 13 13 56 Denmark
