|
BOOKSBOOKSBOOKS
Classification and Learning Using Genetic Algorithms: Applications in Bioinformatics and Web Intelligence
by Sanghamitra Bandyopadhyay and Sankar K. Pal Springer, June 2007
Reviewed by: Zheng Liu (Canada) |
|
A genetic algorithm (GA) is a search technique that can be applied for large, complex, and multimodal search spaces. It emulates biological principles, such as inheritance, mutation, selection, and crossover, to solve complex optimization problems. This 10-chapter book provides a framework describing how GAs can be applied to pattern recognition (PR) and learning systems. Each chapter consists of introduction/background information, theoretical details, experimental results, and a summary. These chapters are well connected through their introductions. Readers can clearly see how the whole book is developed. From the experimental results, readers can also learn how the GA-based method is compared with traditional PR methods. The first chapter gives a brief introduction to pattern recognition, which is good for those who may not have such a background. Chapters 3 to 8 describe the details of the GA algorithm and its use for classification, clustering, and multi-objective optimization. Two specific applications, i.e., learning in bioinformatics and web intelligence, are presented in the last two chapters. The basic principles of GA algorithms are described in Chapter 2. Chapter 3 discusses how the GA algorithm can be applied to one of the "classic problems" in pattern recognition, i.e., supervised classification. GA can be used to facilitate fuzzy rule-based classification and optimize the decision tree method. Moreover, a GA classifier can be created with the training samples by searching for a number of linear segments that form the boundaries between different classes and minimize the misclassification rate. It is interesting to see the performance comparison with some classical methods, such as Bayes maximum likelihood classifier, k-NN (k-nearest neighborhood) classifier, and MLP (multilayer perceptron). It is no surprise that the GA classifier is comparable to, or even better than, those methods. However, the parameter, string length H, is crucial for good performance so needs to be selected carefully. Chapter 4 presents a theoretical analysis of the GA classifier in comparison with the Bayes classifier. For a Bayes classifier, a priori probability and the class conditional density need to be known. However, in a practical application, this may not be possible and this is the gap that a GA classifier can fill. The discussion of the importance of string-length H is continued in Chapter 5. An empirical estimation may degrade the performance of a GA classifier. An automatic evolving process to generate a value of H is described in this chapter. With this value, both the number of misclassified samples and the number of hyperplanes are minimized. The concept of variable length strings in GA is introduced, i.e., the length of the string is not fixed. The name “variable string length genetic algorithm” (VGA) is derived. Chapter 6 describes the integration of variable length chromosomes and GA with chromosome differentiation (GACD), which results in a nonparametric VGACD classifier. The test results of classifying the SPOT image of Calcutta demonstrate the superiority of the VGACD classifier to the VGA classifier, Bayes classifier, and k-NN classifier. In Chapter 7, a multi-objective GA-based classifier is described. Three optimization techniques based on constrained elitist multi-objective GA (CEMOGA), Pareto archived evolutionary strategies (PAES), and non-dominated sorting GA (NSGA-II) are used to develop the multi-objective classifiers. The validating and testing results are presented in the experiments, which indicate the GA-based multi-objective classifier outperforms other multi-objective optimization techniques. Chapter 8 deals with another classical problem in pattern recognition, i.e. clustering or unsupervised classification. Similar to Chapter 3, the authors started with the traditional methods like K-means, and fuzzy c-means clustering. Then, the use of GA to search the appropriate cluster center is described. The details of GA-based approaches for crisp clustering and fuzzy clustering into fixed or variable number of clusters are presented. Two applications using GA-based methods are given in Chapters 9 and 10 respectively. One is bioinformatics and the other is web intelligence. Although few implementation details are provided, readers can learn how to solve a practical problem with GA-based approaches. A flaw in an otherwise perfect book may be some of the figures, which are not uniformly formatted due to the different aspect ratios or to the limitation of paper size. However, this does not hurt the excellent contents presented in the book. This book tries to balance the mixture of theories, algorithms, and applications and is a good reference for people who want to solve a complex optimization problem for their field. As a reader, I may be more curious about how to implement the GA algorithms and how they work for the datasets provided, and I used in this book, even without going through the equations. If the theories were demonstrated with "codes", either commercial or open source software, this would be helpful to a novice. Some information or links about the GA software or the authors' own implementation in the appendix would be an added value and especially helpful to students. Overall, this book is well organized and well written. There is no doubt that this is another good pattern recognition reference to have on one’s bookshelf.
|
|
Click above to go to the publisher’s web page for this book where you will find information about the book, contents and sample pages. |
|
Book Reviews Published in the IAPR Newsletter
Close Range Photogrammetry: Principles, Methods, and Applications by Luhmann, Robson, Kyle, and Harley
Learning Theory: An Approximation Theory Viewpoint by Cucker and Zhou
Character Recognition Systems—A Guide for Students and Practitioners by Cheriet, Kharma, Liu, and Suen
Geometry of Locally Finite Spaces by Kovalevsky
Machine Learning in Document Analysis and Recognition by Marinai and Fujisawa (Editors)
From Gestalt Theory to Image Analysis—A Probabilistic Approach By Desolneux, Moisan, and Morel
Numerical Recipes: The art of scientific computing, 3rd ed. by Press, Teukolsky, Vetterling and Flannery
Feature Extraction and Image Processing, 2nd ed. by Nixon and Aguado
Digital Watermarking and Steganography: Fundamentals and Techniques by Shih
Springer Handbook of Speech Processing by Benesty, Sondhi, and Huang, eds.
Digital Image Processing: An Algorithmic Introduction Using Java by Burger and Burge
Bézier and Splines in Image Processing and Machine Vision by Biswas and Lovell
Practical Algorithms for Image Analysis, 2 ed. by O’Gorman, Sammon and Seul
The Dissimilarity Representation for Pattern Recognition: Foundations and Applications by Pekalska and Duin
Handbook of Biometrics by Jain, Flynn, and Ross (Editors)
Advances in Biometrics – Sensors, Algorithms, and Systems by Ratha and Govindaraju, (Editors)
Dynamic Vision for Perception and Control of Motion by Dickmanns
Bioinformatics by Polanski and Kimmel
Introduction to clustering large and high-dimensional data by Kogan
The Text Mining Handbook by Feldman and Sanger
Information Theory, Inference, and Learning Algorithms by Makay
Geometric Tomography by Gardner
“Foundations and Trends in Computer Graphics and Vision” Curless, Van Gool, and Szeliski., Editors
Applied Combinatorics on Words by M. Lothaire
Human Identification Based on Gait by Nixon, Tan and Chellappar
Mathematics of Digital Images by Stuart Hogan
Advances in Image and Video Segmentation Zhang, Editor
Graph-Theoretic Techniques for Web Content Mining by Schenker, Bunke, Last and Kandel
Handbook of Mathematical Models in Computer Vision by Paragios, Chen, and Faugeras (Editors)
The Geometry of Information Retrieval by van Rijsbergen
Biometric Inverse Problems by Yanushkevich, Stoica, Shmerko and Popel
Correlation Pattern Recognition by Kumar, Mahalanobis, and Juday
Pattern Recognition 3rd Edition by Theodoridis and Koutroumbas
Dictionary of Computer Vision and Image Processing by R.B. Fisher, et. Al
Kernel Methods for Pattern Analysis by Shawe-Taylor and Cristianini
Machine Vision Books
CVonline: an overview
The Guide to Biometrics by Bolle, et al
Pattern Recognition Books Jul. ‘04 [pdf] |