C++ Neural Networks and Fuzzy Logic/Book and Disk [Paperback]

Valluru B. Rao (Author), Hayagriva V. Rao (Author)

Book Description

Publication Date: November 1993

Provides a logical and easy-to-follow presentation of introductory and advanced topics in Neural Network technology. The authors provide numerous examples in C++ for use with most C++ compilers, including Borland and Microsoft. Examples show how to implement neural networks and fuzzy logic for further applications and experimentation.

Product Details

• Paperback: 408 pages
• Publisher: Mis Pr; Pap/Dis edition (November 1993)
• Language: English
• ISBN-10: 155828298X
• ISBN-13: 978-1558282988
• Product Dimensions: 9.2 x 7 x 1.2 inches
• Shipping Weight: 1.6 pounds
• Average Customer Review: 4.0 out of 5 stars  See all reviews (1 customer review)
• Amazon Best Sellers Rank: #1,566,452 in Books (See Top 100 in Books)

More About the Author

Discover books, learn about writers, read author blogs, and more.

Customer Reviews

Most Helpful Customer Reviews

9 of 9 people found the following review helpful:

4.0 out of 5 stars Great Neural Networks Introduction [1993], March 30, 2000

By 

Prof David T Wright (Vancouver, BC) - See all my reviews

This review is from: C++ Neural Networks and Fuzzy Logic/Book and Disk (Paperback)

Despite being used for decades controlling safety/time-critical non-linear manufacturing processes (and other applications), the neural networks/AI label evokes images from science fiction by the general media. Beyond the initially intimidating title, this book is of real use to researchers addressing complex problems, and those trying to use neural networks. Without resorting to hardwired neural networks, or MATLAB with neural networks toolbox, one can readily understand and use the included (dated now) C++ neural network code (C compiler excluded- but use Borland/Microsoft).

The basic neural network steps include: model process (e.g. manufacturing), gather data, pre-process data, compile, link & run the neural network code, let it iterate & learn, analyse output data sets, then use to hard-code a PID-controller (say). Even a 486-66/Win3.1/16MB RAM CPU can readily handle 25 variable/1000+ data point (pultrusion) manufacturing process (1994). As I often tend to use fuzzy-pre processing, and C/C++ for hacking demonstrators, the combined treatment of fuzzy logic, C++ and neural networks works very well.

Topics include: fuzzy logic introduction, constructing a neural network, C++ and object orientation, models and Adaptive Resonance Theory (ART), learning, self organization & resonance, backpropogation and non-linear optimization, Bidirectional Associative Memory (BAM), Fuzzy Associative Memory (FAM), and applications for financial modelling.

Weaknesses include: the now dated and programming-novice unfriendly software, and the ultimately limited software- far better when you are experienced to use MATLAB with neural networks toolbox for developing fast, usable networks.

Overall, the style is approachable, and the content readily understandable and usable by the typical professional engineer/graduate engineering student audience. This book includes much helpful annotated pseudo-code, examples, references, defined terms, and mathematical explanations.