On the Detection of Conflicts in Diagnostic Bayesian Networks Using Abstraction
This work addresses the brittleness problem for users of expert systems, but it is incremental as it builds on existing straw model methods.
The paper tackles the brittleness problem in expert systems by developing an algorithm that automatically constructs a bipartite straw model from a diagnostic Bayesian network to detect conflicts. The result shows that in some cases, this straw model outperforms the independent straw model of Jensen et al., though no concrete numbers are provided.
An important issue in the use of expert systems is the so-called brittleness problem. Expert systems model only a limited part of the world. While the explicit management of uncertainty in expert systems itigates the brittleness problem, it is still possible for a system to be used, unwittingly, in ways that the system is not prepared to address. Such a situation may be detected by the method of straw models, first presented by Jensen et al. [1990] and later generalized and justified by Laskey [1991]. We describe an algorithm, which we have implemented, that takes as input an annotated diagnostic Bayesian network (the base model) and constructs, without assistance, a bipartite network to be used as a straw model. We show that in some cases this straw model is better that the independent straw model of Jensen et al., the only other straw model for which a construction algorithm has been designed and implemented.