Fuzzy rule-based expert systems and genetic machine learning. 2nd rev. and enlarged ed. (English) Zbl 0914.68166

Studies in Fuzziness. 3. Heidelberg: Physica-Verlag. xx, 432 p. (1996).
This monograph presents fuzzy rule-based expert systems, and shows how they can be integrated with genetic algorithms and neural networks to provide tools for automatic knowledge acquisition and machine learning. The book is divided into two parts. In Part 1, fuzzy rule-based expert systems and their foundations are presented. A brief introduction to fuzzy set theory, and a review of fuzzy systems research is provided. The main ideas of fuzzy thinking including the boundary problem, the principle of in compatibility, the principle of computational efficiency, and the principle of reasoning in spite of inconsistency, are discussed. To present natural language computation, a small fragment of natural language sufficient Part 2 describes fuzzy classifier systems. These are a combination of fuzzy rule-languages and genetic algorithms, and are intended to learn fuzzy production rules. A short introduction to genetic algorithms is given, and it is shown how variants of genetic algorithms can be developed.


68T05 Learning and adaptive systems in artificial intelligence
68T35 Theory of languages and software systems (knowledge-based systems, expert systems, etc.) for artificial intelligence
68-02 Research exposition (monographs, survey articles) pertaining to computer science