Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
Automatic refinement of expert system knowledge bases
Automatic refinement of expert system knowledge bases
More results on the complexity of knowledge base refinement: belief networks
Proceedings of the seventh international conference (1990) on Machine learning
Knowledge base refinement and theory revision
Proceedings of the sixth international workshop on Machine learning
Some results on the computational complexity of refining confidence factors
International Journal of Approximate Reasoning
On the conversion of rule bases into belief networks
SAC '92 Proceedings of the 1992 ACM/SIGAPP Symposium on Applied computing: technological challenges of the 1990's
Inductive Inference: Theory and Methods
ACM Computing Surveys (CSUR)
Computational Complexity of Machine Learning
Computational Complexity of Machine Learning
Computers and Intractability: A Guide to the Theory of NP-Completeness
Computers and Intractability: A Guide to the Theory of NP-Completeness
Explanation-Based Generalization: A Unifying View
Machine Learning
Inductive learning from good examples
IJCAI'91 Proceedings of the 12th international joint conference on Artificial intelligence - Volume 2
Optimal refinement of rule bases
AI Communications
Optimal refinement of rule bases
AI Communications
On automatic knowledge validation for Bayesian knowledge bases
Data & Knowledge Engineering
The complexity of theory revision
Artificial Intelligence
| Hi-index | 0.00 |
Refining deep (multilayer) rule bases of an expert system with uncertainty to cover a set of new examples can be very difficult (NP-hard). We analyze refinement via reduction, an approach first proposed by Ginsberg, who claimed that this approach eases the complexity of refining rule bases without uncertainty. We outline a model of rule bases with uncertainty, and give necessary and sufficient conditions on uncertainty combination functions that permit reduction from deep to flat (nonchaining) rule bases. We prove that reduction cannot be performed with most commonly used uncertainty combination functions. However, we show that there is a class of reducible rule bases in which the strength refinement problem is NP-hard in the deep rule base, reduction is polynomial, and the flat rule base can be refined in polynomial time. This result also allows polynomial refinement of practical expert systems in the form of rule deletion. Thus, our results provide some theoretical evidence that refinement via reduction is feasible.