A Generalization of the Noisy-Or Model
This work provides a useful modeling aid for constructing Bayesian networks in domains such as engineering and reliability analysis, but it is incremental as it builds directly on an existing model.
The paper tackles the problem of extending the Noisy-Or model from Boolean variables to n-ary inputs and outputs and arbitrary functions, resulting in a generalization that aids in constructing Bayesian networks for applications like digital circuit diagnosis and network reliability analysis.
The Noisy-Or model is convenient for describing a class of uncertain relationships in Bayesian networks [Pearl 1988]. Pearl describes the Noisy-Or model for Boolean variables. Here we generalize the model to nary input and output variables and to arbitrary functions other than the Boolean OR function. This generalization is a useful modeling aid for construction of Bayesian networks. We illustrate with some examples including digital circuit diagnosis and network reliability analysis.