Probabilistic Expert Systems

by Glenn Shafer

Probabilistic Expert Systems emphasizes the basic computational principles that make probabilistic reasoning feasible in expert systems. The key to computation in these systems is the modularity of the probabilistic model. Shafer describes and compares the principal architectures for exploiting this modularity in the computation of prior and posterior probabilities. He also indicates how these similar yet different architectures apply to a wide variety of other problems of recursive computation in applied mathematics and operations research. This book describes probabilistic expert systems in a more rigorous and focused way than existing literature, and provides an annotated bibliography that includes pointers to conferences and software. Also included are exercises that will help the reader begin to explore the problem of generalizing from probability to broader domains of recursive computation.

88 pages, Paperback

First published January 1, 1987

Reviews

[Kevin K. Gillette · Rating 4 out of 5]

Professor Shafer's contribution to the CBMS series, though written nearly 20 years ago (1996), stands the test of time in the way in which it matches up with more recent versions of the ideas. A modern adjunct to this little volume is the much more ponderous "Probabilistic Graphical Models" by Daphne Koller (reviewed by this writer earlier).

The idea is basically the same: To attach a probabilistic framework to decision trees, e.g. in examining information prior to making a preliminary medical diagnosis, or selection a recommendation for a customer based on past purchases. What's especially interesting about this book is that, although it is part of a series of monographs that are (generally) fairly sophisticated, the mathematics here is comparatively elementary: an undergraduate with a basic course in probability theory, and a very rudimentary understanding of graph theory (vertices and edges), along with a soupcon of patience for notation can easily follow the reasoning herein.

I recommend this book for anyone interested in the area of fuzzy logic (in the computer science sense).