Metaheuristic Clustering

by Das, Swagatam; Abraham, Ajith; Konar, Amit

• ISBN13: 9783540921721
• ISBN10: 3540921729
• Format: Hardcover
• Copyright: 2009-04-01
• Publisher: Springer Verlag

Summary

The series Studies in Computational Intelligence (SCI) publishes new developments and advances in the various areas of computational intelligence-quickly and with a high quality. The intent is to cover the theory, applications, and design methods of computational intelligence, as embedded in the fields of engineering, computer science, physics and life science, as well as the methodologies behind them. The series contains monographs, lecture notes and edited volumes in computational intelligence spanning the areas of neural networks, connectionist systems, genetic algorithms, evolutionary computation, artificial intelligence, cellular automata, self-organizing systems, soft computing, fuzzy systems and hybrid intelligent systems. Critical to both contributors and readers are the short publication time and world-wide distribution-this permits a rapid and broad dissemination of research results.

Table of Contents

Metaheuristic Pattern Clustering-An Overview	p. 1
Introduction	p. 1
The Clustering Problem	p. 6
Basic Definitions	p. 6
Proximity Measures	p. 8
Clustering Validity Indices	p. 9
The Davis-Bouldin (DB) Index	p. 9
The Dunn and Dunn Like Indices	p. 10
S_Dbw Validity Index	p. 10
Partition Coefficient	p. 11
Classification Entropy	p. 12
Xie-Beni Index	p. 12
The PS Measure	p. 12
The PBMF Index	p. 13
The CS Measure	p. 13
The Classical Clustering Algorithms	p. 14
Hierarchical Clustering Algorithms	p. 14
Partitional Clustering Algorithms	p. 16
The k-Means Algorithm	p. 18
The k-Medoids Algorithm	p. 19
The Fuzzy c-Means Algorithm	p. 19
The Expectation-Maximization Algorithm	p. 20
The k-Harmonic Means Algorithm	p. 21
Density-Based Clustering Algorithms	p. 22
Grid-Based Clustering Algorithms	p. 23
A Comparative View of the Traditional Clustering Algorithms	p. 23
Population Based Optimization Techniques	p. 26
Optimization Algorithms	p. 26
The Evolutionary Computing (EC) Family	p. 28
The Evolutionary Algorithms	p. 29
Evolutionary Strategies (ESs)	p. 30
Evolutionary Programming (EP)	p. 30
Genetic Algorithms (GAs)	p. 31
Genetic Programming (GPs)	p. 33
Swarm Intelligence Algorithms	p. 33
The Particle Swarm Optimization (PSO)	p. 34
The Ant Colony Optimization (ACO)	p. 35
Evolutionary Computing (EC) Techniques in Pattern Clustering	p. 36
Clustering Methods Based on Evolutionary Algorithms	p. 36
The GA-Based Partitional Clustering Algorithms-Earlier Approaches	p. 37
Clustering Algorithms Based on ES, EP, and GP	p. 38
Clustering Using Swarm Intelligence Algorithms	p. 39
The Ant Colony Based Clustering Algorithms	p. 39
The PSO-Based Clustering Algorithms	p. 40
Automatic Clustering: Evolutionary Vs. Classical Approaches	p. 42
Genetic Clustering with Unknown Number of Clusters K (GCUK) Algorithm	p. 43
The FVGA Algorithm	p. 44
The Dynamic Clustering with Particle Swarm Optimization Algorithm	p. 45
Clustering with Evolutionary Multi-objective Optimization	p. 45
Multi-objective Optimization Problem (MOP)	p. 45
Evolutionary Multi-objective Optimization (EMO)	p. 46
Clustering Using EMO Algorithms (EMOAs)	p. 48
Innovation and Research: Main Contributions of This Volume	p. 49
Conclusions	p. 53
References	p. 53
Differential Evolution Algorithm: Foundations and Perspectives	p. 63
Introduction	p. 63
Differential Evolution: A First Glance	p. 64
Initialization of the Parameter Vectors	p. 64
Mutation with Differential Operators	p. 66
Crossover	p. 68
Selection	p. 72
Summary of DE Iteration	p. 73
The Complete Differential Evolution Algorithm Family of Storn and Price	p. 77
Control Parameters of the Differential Evolution	p. 79
Important Variants of the Differential Evolution Algorithm	p. 81
Differential Evolution Using Trigonometric Mutation	p. 81
Differential Evolution Using Arithmetic Recombination	p. 82
Self Adaptive Differential Evolution	p. 84
The DE/rand/1/Either-Or Algorithm	p. 86
The Opposition-Based Differential Evolution	p. 86
The Binary Differential Evolution Algorithm	p. 89
Differential Evolution with Adaptive Local Search	p. 90
Self-adaptive Differential Evolution (SaDE) with Strategy Adaptation	p. 92
DE with Neighborhood-Based Mutation	p. 93
The DE/target-to-best/1 - A Few Drawbacks	p. 93
Motivations for the Neighborhood-Based Mutation	p. 94
The Local and Global Neighborhood-Based Mutations in DE	p. 95
Control Parameters in DEGL	p. 97
Runtime Complexity of DEGL - A Discussion	p. 99
Comparative Performance of DEGL	p. 102
Hybridization of Differential Evolution with Other Stochastic Search Techniques	p. 104
Conclusions	p. 106
References	p. 107
Modeling and Analysis of the Population-Dynamics of Differential Evolution Algorithm	p. 111
Introduction	p. 111
The Mathematical Model of the Population-Dynamics in DE	p. 112
Assumptions	p. 113
Modeling Different Steps of DE	p. 114
A State Space Formulation of the DE Population	p. 122
Lyapunov Stability Analysis of the DE Population	p. 124
Computer Simulation Results	p. 129
Conclusions	p. 131
Appendix	p. 132
References	p. 133
Automatic Hard Clustering Using Improved Differential Evolution Algorithm	p. 137
Introduction	p. 137
The DE-Based Automatic Clustering Algorithm	p. 138
Vector Representation	p. 138
Designing the Fitness Function	p. 140
Avoiding Erroneous Vectors	p. 146
Modification of the Classical DE	p. 147
Pseudo-code of the ACDE Algorithm	p. 148
Experiments and Results for Real Life Datasets	p. 148
The Datasets Used	p. 149
Population Initialization	p. 149
Parameter Setup for the Algorithms Compared	p. 150
Simulation Strategy	p. 150
Empirical Results	p. 151
Discussion on the Results (for Real Life Datasets)	p. 161
Application to Image Segmentation	p. 162
Image Segmentation as a Clustering Problem	p. 162
Experimental Details and Results	p. 162
Discussion on Image Segmentation Results	p. 165
Conclusions	p. 172
Appendix: Statistical Tests Used	p. 172
References	p. 173
Fuzzy Clustering in the Kernel-Induced Feature Space Using Differential Evolution Algorithm	p. 175
Introduction	p. 175
The Kernel-Induced Clustering	p. 177
The Kernel-Induced Clustering Technique with DEGL	p. 181
Kernelization of the Xie-Beni Index	p. 181
Summary of the Integrated Clustering Approach	p. 183
Experimental Results	p. 184
General Comparison with Other Clustering Algorithms	p. 184
Scalability of the DEGL-Based Clustering Algorithm	p. 194
Application to Image Pixel Clustering	p. 197
Parametric Setup for the Contestant Algorithms	p. 197
The Test-Suite for Comparison	p. 198
Quantitative Validation of Clustering Results	p. 198
The Simulation Strategy	p. 199
Experimental Results	p. 200
Discussion on the Results	p. 202
Conclusions	p. 208
References	p. 208
Clustering Using Multi-objective Differential Evolution Algorithms	p. 213
Introduction	p. 213
Multi-objective Optimization Using Differential Evolution Algorithm	p. 215
The Pareto Differential Evolution (PDE)	p. 215
The Multi-Objective Differential Evolution (MODE)	p. 216
Differential Evolution for Multi-objective Optimization (DEMO)	p. 216
Non-dominated Sorting DE (NSDE)	p. 218
The Multi-objective Clustering Scheme	p. 218
Search-Variable Representation	p. 218
Selecting the Objective Functions	p. 219
Selecting the Best Solutions from Pareto-front	p. 221
Evaluating the Clustering Quality	p. 222
Experiments and Results	p. 223
Datasets Used	p. 223
Parameters for the Algorithms	p. 223
Presentation of Results	p. 224
Significance and Validation of Microarray Data Clustering Results	p. 228
Conclusions	p. 236
References	p. 237
Conclusions and Future Research	p. 239
Cluster Analysis Using Metaheuristics: A Roadmap of This Volume	p. 239
Potential Application Areas for Clustering Schemes	p. 241
Future Research Directions	p. 242
References	p. 245
Index	p. 249
Table of Contents provided by Ingram. All Rights Reserved.