Soft Computing and Intelligent Systems: Theory and Applications

Citation

Harwood, C.J. (2001), "Soft Computing and Intelligent Systems: Theory and Applications", Kybernetes, Vol. 30 No. 1, pp. 103-115. https://doi.org/10.1108/k.2001.30.1.103.3

Both the terms “soft computing” and “intelligent systems” need to be defined or at least discussed at some length. Using them in the title may or, indeed, may not attract the casual reader. Delving more deeply into the book we do get an impression of what the authors are aiming at presenting. The days of selling a book simply by having “computing” or “intelligence” in the title are long gone. Scientists and particularly cyberneticians, who number computer scientists in their ranks, are now much more discerning.

One interpretation of “soft computing” is that it is about exploiting the tolerance for imprecision and uncertainty in the solutions offered to problems arising in information and intelligent systems. That in itself is a massive challenge to any author(s). In this project they have tackled it by calling on researched contributions in the subject and presenting them in a classified format. They have centred on: foundations; theory; implications; applications; and future perspectives. Casting the contributions into just five sections has obvious disadvantages – nothing is usually black or white, as the methodologies they have chosen to present confirm. They believe that the most important methodologies that make up “soft computing” are: fuzzy logic; neurocomputing; evolutionary computing; probabilistic computing; and machine learning. The book deals with these in what is a lengthy text. But the individual methodologies are themselves already the subject of extensive coverage in the literature. This in itself made for difficulties, and at times it was not easy to gauge the audience that the authors were concerned with.

The subject was, however, competently dealt with and will undoubtedly attract many readers to learn more about soft computing and, indeed, apply some of the methodologies discussed.