1 of 1 people found the following review helpful:
5.0 out of 5 stars
A must-have for beginners and experts, July 29, 1999
By A Customer
I have worked with knowledge engineering for 15 years and I still find this the best introductory book about expert systems.
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No
3.0 out of 5 stars
Expert Systems as They Were, December 4, 2010
This review is from: Expert Systems: Artificial Intelligence in Business (General Trade) (Paperback)
Written by management consultant Paul Harmon and psychologist David King, this is not the typical technical introduction to an artificial intelligence topic. It covers the basic concepts of rule-based expert systems sufficiently well to prepare readers to use a simple development tool. Its strength is the added perspective on modeling expert knowledge and implementing systems in a business environment.
The book's sixteen chapters are divided into four sections. The first section introduces basic concepts and techniques of expert system development. An initial historical review focuses on the MYCIN medical diagnosis system. Readers then learn some basic principles of human problem solving. Then the focus shifts to how computer tools represent expert knowledge and use it to draw inferences from facts entered by a user.
The second section reviews AI programming languages and higher-level development tools. The historical information remains interesting and the introductory presentation of the LISP and prolog languages is still relevant. Unfortunately the comparative review of commercial tools is now badly out of date. The fourth section--I am jumping ahead--discusses expert systems in specialized domains such as medicine and equipment troubleshooting. The chapters which predict the future--Knowledge Systems in the Next Five Years and Preparing for the Knowledge Systems Revolution--are not only dated, but a little depressing to read. The revolution never came; expert systems disappeared into database business rules and the "wizards" in software help systems.
Section three, however, contains still-useful guidance about how to observe and interview a human expert and build a rule-based model of their problem-solving expertise. The example expert system models a decision about what media (book, video, etc.) to use for different training tasks. This system is just the right size--large enough to exhibit some complexity, but small enough to be understandable. It sets the stage for the authors' advice about managing larger expert system development projects.
I recommend the book for those interested in the expert systems boom of the mid-1980s and in the technology it was based on. Those wishing a more current introduction should try Peter Jackson's
Introduction to Expert Systems. Giarratano and Riley's
Expert Systems: Principles and Programming is a very good technical text build around the public domain CLIPS system.
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No