##
**Probabilistic expert systems.**
*(English)*
Zbl 0866.68108

CBMS-NSF Regional Conference Series in Applied Mathematics. 67. Philadelphia, PA: SIAM, Society for Industrial and Applied Mathematics. viii, 80 p. (1996).

This monograph represents the revised lectures on Regional Conference at the North Dakota University in 1992.

The author analyzes join-tree methods for the computation of prior and posterior probabilities in belief nets which continue to be central in the theory and practice of probabilistic expert systems. He discusses the question how the basic architectures for join-tree computation can be applied to the other methods for combining evidence (to the belief-function method) and to various problems of applied mathematics and operations research. Many aspects of computation in expert systems especially Markov chain Monte Carlo approximation, computation for model selection and computation for model evaluation are presented here. In comparison with North Dakota lectures the brief chapter of annotated bibliography and some useful exercises are added.

The reviewing monograph should be useful to scholars and students in artificial intelligence, operations research and the various branches of applied statistics that use probabilistic methods.

The author analyzes join-tree methods for the computation of prior and posterior probabilities in belief nets which continue to be central in the theory and practice of probabilistic expert systems. He discusses the question how the basic architectures for join-tree computation can be applied to the other methods for combining evidence (to the belief-function method) and to various problems of applied mathematics and operations research. Many aspects of computation in expert systems especially Markov chain Monte Carlo approximation, computation for model selection and computation for model evaluation are presented here. In comparison with North Dakota lectures the brief chapter of annotated bibliography and some useful exercises are added.

The reviewing monograph should be useful to scholars and students in artificial intelligence, operations research and the various branches of applied statistics that use probabilistic methods.

Reviewer: S.G.Valeev