Computing Reviews

Computational intelligence: a methodological introduction (3rd ed.)
Kruse R., Mostaghim S., Borgelt C., Braune C., Steinbrecher M., Springer International Publishing,Cham, Switzerland,2022. 653 pp.Type:Book
Date Reviewed: 03/07/23

Ideas, more than inventions, change the world. Seen from this perspective, computational intelligence should be viewed, in my opinion, as a set of ideas that find applications in intelligent systems or devices that exhibit intelligent behaviour in some sense (such as the computer). The algorithms, concepts, and techniques that come within the periphery of computational intelligence are mostly based on natural phenomenon, or, in other words, the modus operandi of nature inspires the strategies of computational intelligence. This introductory text by Kruse et. al. is a nice blend of theory and applications on the topic.

Here are some salient features of the book:

  • State of the art is lucidly covered, including topics that have been developed recently like probabilistic graphical models;
  • Traditional computational intelligence topics are discussed at length, including artificial neural networks (ANNs), evolutionary algorithms, Bayesian and Markov networks, and fuzzy systems;
  • Sufficient new material on topics like deep learning, large-scale optimization, scalarization, and collective decision-making algorithms;
  • Many exercises and solutions that help readers grasp the concepts; and
  • Useful additional materials through software tools, websites, and lecture slides.

After a brief introduction in chapter 1, the authors proceed to cover neural networks quite comprehensively in Part 1. Chapters 2 through 10 look at ANNs, threshold logic units, graph neural networks (GNNs), multi-layer perceptions, radial basis function networks, self-organizing maps, Hopfield networks, recurrent networks, and mathematical remarks in neural networks, respectively. Part 2 details evolutionary algorithms, including genetic algorithms and swarm intelligence (chapters 11 to 14). Parts 3 and 4, “Fuzzy Systems” and “Bayesian Networks,” deal with the extension principle, fuzzy and similarity relations, approximate reasoning, and fuzzy data analysis (chapters 15 to 22), and probability, graph theory, decomposition, propagation, and causal networks (chapters 23 to 30).

The authors have written Computational intelligence in such a way that it can serve as both a textbook and a helpful reference book for students and practitioners of computing science and related fields. The presentation is careful and friendly yet technically sound.

Reviewer:  Soubhik Chakraborty Review #: CR147559

Reproduction in whole or in part without permission is prohibited.   Copyright 2023™
Terms of Use
| Privacy Policy