Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Multidimensional particle swarm optimization for machine learning and pattern recognition
Kiranyaz S., Ince T., Gabbouj M., Springer Publishing Company, Incorporated, New York, NY, 2013. 270 pp. Type: Book (978-3-642378-45-4)
Date Reviewed: Jul 30 2014

Recent advances in optimization, namely the optimization technique called multidimensional particle swarm optimization, are explored in this book.

While chapter 1 is not particularly detailed, it provides a comprehensive overview of optimization techniques. The chapter is well structured and is divided into three parts, covering deterministic methods, stochastic methods, and evolutionary algorithms. These represent the three main directions in the development of optimization techniques. The most interesting part is Subsection 2.4.2 on differential evolution, which is a modern technique similar to genetic algorithms but applicable to real-valued vectors rather than bit-encoded strings. The section is very short and I believe a more extensive discussion would have benefitted readers.

Chapters 3 and 4 introduce in detail particle swarm optimization, which was introduced in 1995 with the goal to converge to the global optima of some multidimensional and/or possibly nonlinear systems. The most interesting part of chapter 3 is on applications (Subsections 3.4.1 and 3.4.2), where nonlinear minimization and data clustering in a particle swarm optimization framework are described. Subsection 3.4.3 describes artificial neural networks with examples in medicine.

A significant plus of the book is the clearly organized programming mini-guide given in Subsection 3.5. Dynamic data clustering via the multidimensional particle swarm optimization approach, introduced briefly in Subsection 3.4.2 as an application example, is expanded and discussed in detail in chapter 6. The chapter, however, is based on the fractional global best formation (FGBF) concept discussed and explained in chapter 5. As such, those interested in clustering must first familiarize themselves with the FGBF. Thus, chapter 5, “Improving Global Convergence,” is a foundation for material in chapters 6 and 7. Some of the most interesting parts in chapter 5 are the performance evaluation paragraphs (Subsections 5.5.1 and 5.5.2).

The next three chapters (8, 9, and 10) have strong practical value and are filled with many useful considerations. The information presented can give developers and graduate students in the field of optimization a jump-start in code development. The application chapters provide well-commented source code, which can be used directly or as a template in development.

This book has enough material to be used as a reference text in research in areas of biomedical signal processing, classification, and clustering. Alternatively, it can be employed as an extra textbook in a graduate course on optimization. Its clear style and strong practical orientation make the book an excellent addition to the bookshelf of any researcher dealing with optimization problems in many dimensions.

Reviewer:  Alexander Tzanov Review #: CR142568 (1411-0940)
Bookmark and Share
  Reviewer Selected
Featured Reviewer
 
 
Learning (I.2.6 )
 
 
Applications And Expert Systems (I.2.1 )
 
 
Optimization (G.1.6 )
 
 
Image Processing And Computer Vision (I.4 )
 
 
Pattern Recognition (I.5 )
 
Would you recommend this review?
yes
no
Other reviews under "Learning": Date
Learning in parallel networks: simulating learning in a probabilistic system
Hinton G. (ed) BYTE 10(4): 265-273, 1985. Type: Article
Nov 1 1985
Macro-operators: a weak method for learning
Korf R. Artificial Intelligence 26(1): 35-77, 1985. Type: Article
Feb 1 1986
Inferring (mal) rules from pupils’ protocols
Sleeman D.  Progress in artificial intelligence (, Orsay, France,391985. Type: Proceedings
Dec 1 1985
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy