Mumford Computational Intelligence

Collaboration, Fusion and Emergence
1. Auflage 2009
ISBN: 978-3-642-01799-5
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, Band 1, 732 Seiten

Reihe: Intelligent Systems Reference Library

This book is about synergy in computational intelligence (CI). It is a c- lection of chapters that covers a rich and diverse variety of computer-based techniques, all involving some aspect of computational intelligence, but each one taking a somewhat pragmatic view. Many complex problems in the real world require the application of some form of what we loosely call “intel- gence”fortheirsolution. Fewcanbesolvedbythenaiveapplicationofasingle technique, however good it is. Authors in this collection recognize the li- tations of individual paradigms, and propose some practical and novel ways in which di?erent CI techniques can be combined with each other, or with more traditional computational techniques, to produce powerful probl- solving environments which exhibit synergy, i. e. , systems in which the whole 1 is greater than the sum of the parts . Computational intelligence is a relatively new term, and there is some d- agreement as to its precise de?nition. Some practitioners limit its scope to schemes involving evolutionary algorithms, neural networks, fuzzy logic, or hybrids of these. For others, the de?nition is a little more ?exible, and will include paradigms such as Bayesian belief networks, multi-agent systems, case-based reasoning and so on. Generally, the term has a similar meaning to the well-known phrase “Arti?cial Intelligence” (AI), although CI is p- ceived moreas a “bottom up” approachfrom which intelligent behaviour can emerge,whereasAItendstobestudiedfromthe“topdown”,andderivefrom pondering upon the “meaning of intelligence”. (These and other key issues will be discussed in more detail in Chapter 1.

Autoren/Hrsg.

Mumford, Christine L.

Inhaltsverzeichnis

1;Title Page;2
2;Preface;8
3;Contents;11
4;Part I Introduction;14
4.1;Synergy in Computational Intelligence;15
4.1.1;Introduction;15
4.1.2;The Birth of Computational Intelligence;16
4.1.3;The Main CI Techniques;18
4.1.3.1;Evolutionary Algorithms;18
4.1.3.2;Neural Networks;20
4.1.3.3;Fuzzy Systems;21
4.1.3.4;Multi-Agent Systems;23
4.1.3.5;Other Techniques Covered in the Book;24
4.1.4;Chapters Included in This Book;24
4.1.4.1;Part I: Introduction;25
4.1.4.2;Part II: Fusing Evolutionary Algorithms and Fuzzy Logic;25
4.1.4.3;Part III: Adaptive Solution Schemes;26
4.1.4.4;Part IV: Multi-Agent Systems;27
4.1.4.5;Part V: Computer Vision;28
4.1.4.6;Part VI: Communication for CI Systems;29
4.1.4.7;Part VII: Artificial Immune Systems;30
4.1.4.8;Part VIII: Parallel Evolutionary Algorithms;30
4.1.4.9;Part IX: CI for Clustering and Classification;31
4.1.5;References;33
4.2;Computational Intelligence: The Legacy of Alan Turing and John von Neumann;34
4.2.1;Introduction;34
4.2.2;Turing and Machine Intelligence;36
4.2.2.1;Turing’s Construction of an Intelligent Machine;37
4.2.2.2;Turing on Learning and Evolution;38
4.2.2.3;Turing and Neural Networks;39
4.2.2.4;Discipline and Initiative;40
4.2.3;Von Neumann’s Logical Theory of Automata;41
4.2.3.1;McCulloch-Pitts Theory of Formal Neural Networks;42
4.2.3.2;Complication and Self-reproduction;43
4.2.4;Holland’s Logical Theory of Adaptive Systems;44
4.2.5;The Beginning of Artificial Intelligence - The Logic Theorist;46
4.2.6;Discussion of the Early Proposals to Create Artificial Intelligence by Simulating Evolution;47
4.2.7;Cyc and Cog: Two Large Projects in the Legacy of Alan Turing;49
4.2.7.1;The Cyc Project;49
4.2.7.2;The Cog Project;50
4.2.8;The JANUS Hand-Eye Robot and the Pandemonium Architecture;51
4.2.9;Conclusion;52
4.2.10;References;53
5;Part II Fusing Evolutionary Algorithms and Fuzzy Logic;55
5.1;Multiobjective Evolutionary Algorithms for Electric Power Dispatch Problem;56
5.1.1;Introduction;56
5.1.2;EED Problem Formulation;59
5.1.2.1;Problem Objectives;59
5.1.2.2;Problem Constraints;60
5.1.2.3;Problem Formulation;61
5.1.3;Multiobjective Optimization;61
5.1.3.1;Principles and Definitions;61
5.1.3.2;Fitness Assignment;62
5.1.3.3;Diversity Preservation;63
5.1.4;Multiobjective Evolutionary Algorithms;63
5.1.4.1;Non-dominated Sorted Genetic Algorithm (NSGA);63
5.1.4.2;Niched Pareto Genetic Algorithm (NPGA);65
5.1.4.3;Strength Pareto Evolutionary Algorithm (SPEA);67
5.1.5;MOEA Implementation;68
5.1.5.1;Reducing the Pareto Set by Clustering;68
5.1.5.2;Best Compromise Solution;69
5.1.5.3;Real-Coded Genetic Algorithm;70
5.1.5.4;The Computational Flow;71
5.1.5.5;Settings of the Proposed Approach;71
5.1.6;Results and Discussions;74
5.1.7;A Comparative Study;82
5.1.8;Future Work;88
5.1.9;Conclusions;88
5.1.10;References;89
5.2;Fuzzy Evolutionary Algorithms and Genetic Fuzzy Systems: A Positive Collaboration between Evolutionary Algorithms and Fuzzy Systems;92
5.2.1;Introduction;92
5.2.2;Fuzzy Rule Based Systems;93
5.2.2.1;Preliminaries: Fuzzy Set and Linguistic Variable;94
5.2.2.2;Basic Elements of FRBSs;96
5.2.2.3;Example of FRBS: Fuzzy Logic Control of an Inverted Pendulum;98
5.2.3;Genetic Fuzzy Systems;100
5.2.3.1;Taxonomy of Genetic Fuzzy Systems;101
5.2.3.2;Genetic Learning: Rule Coding and Cooperation/Competition Evolutionary Process;106
5.2.3.3;Some GFS Milestones: Books and Special Issues;107
5.2.3.4;Current Research Trends in GFSs;110
5.2.4;Fuzzy Evolutionary Algorithms;114
5.2.4.1;Fuzzy Adaptive GAs;114
5.2.4.2;EA Components Based on Fuzzy Tools;119
5.2.4.3;Other Fuzzy EA Models;122
5.2.4.4;Future Work on Fuzzy EAs;125
5.2.5;Concluding Remarks;128
5.2.6;References;128
5.3;Multiobjective Genetic Fuzzy Systems;140
5.3.1;Introduction;140
5.3.2;Evolutionary Multiobjective Optimization;145
5.3.2.1;Some Basic Concepts in Multiobjective Optimization;145
5.3.2.2;Evolutionary Multiobjective Optimization;146
5.3.3;Multiobjective Genetic Fuzzy Systems;149
5.3.4;Multiobjective Genetic Fuzzy Rule Selection;151
5.3.4.1;Fuzzy Rule-Based Classifiers;151
5.3.4.2;Candidate Rule Generation;153
5.3.4.3;Multiobjective Fuzzy Rule Selection;154
5.3.4.4;Computational Experiments;155
5.3.5;Multiobjective Fuzzy Genetics-Based Machine Learning;163
5.3.5.1;Two Approaches in Genetics-Based Machine Learning;163
5.3.5.2;Implementation of Multiobjective Fuzzy GBML Algorithms;163
5.3.5.3;Computational Experiments;166
5.3.6;Related Studies;170
5.3.6.1;Evolutionary Multiobjective Data Mining;170
5.3.6.2;Evolutionary Multiobjective Feature Selection;171
5.3.6.3;Evolutionary Multiobjective Clustering;171
5.3.6.4;Evolutionary Ensemble Design;172
5.3.6.5;Evolutionary Multiobjective Neural Network Design;172
5.3.6.6;Multiobjective Genetic Programming;172
5.3.7;Future Research Directions;172
5.3.8;Conclusions;174
5.3.9;References;174
6;Part III Adaptive Solution Schemes;183
6.1;Exploring Hyper-heuristic Methodologies with Genetic Programming;184
6.1.1;Introduction;184
6.1.1.1;The Need for Heuristics;185
6.1.1.2;Hyper-heuristics;185
6.1.1.3;Genetic Programming;186
6.1.1.4;Chapter Outline;189
6.1.2;Genetic Programming as a Hyper-heuristic;189
6.1.2.1;Suitability of Genetic Programming as a Hyper-heuristic;189
6.1.2.2;The Basic Approach;190
6.1.3;Case Studies;190
6.1.3.1;Boolean Satisfiability – SAT;191
6.1.3.2;Online Bin Packing;194
6.1.4;Literature Review;199
6.1.4.1;Genetic Programming Hyper-heuristics for Generating Reusable Heuristics;199
6.1.4.2;Genetic Programming Hyper-heuristics for Generating Disposable Heuristics;202
6.1.4.3;Learning to Learn;202
6.1.4.4;The Teacher System;203
6.1.4.5;Related Areas;203
6.1.5;Summary and Conclusion;204
6.1.5.1;The Need for Automatic Heuristic Generation;204
6.1.5.2;Supplementing Human Designed Heuristics;204
6.1.6;References;205
6.2;Adaptive Constraint Satisfaction: The Quickest First Principle;209
6.2.1;Introduction;209
6.2.2;The Adaptive Strategy;211
6.2.2.1;Chain Design;211
6.2.2.2;Switching Policy;212
6.2.3;The Reduced Exceptional Behaviour Algorithm (REBA);213
6.2.3.1;The REBA Algorithm Chain;213
6.2.3.2;The Monitor Search Level (MSL) Thrashing Predictor;214
6.2.3.3;The REBA Switching Mechanism;218
6.2.4;Experiments;218
6.2.4.1;Experimental Design;218
6.2.4.2;The Effectiveness of REBA;220
6.2.4.3;Evaluation of the MSL Predictor;222
6.2.5;Discussion;225
6.2.6;Appendix;227
6.2.6.1;A.1 Tables of results for Figures 4–7;227
6.2.6.2;A.2 Results for 100 Variables;229
6.2.6.3;A.3 Tables of results for Figures 14 – 17;231
6.2.7;References;235
7;Part IV Multi-Agent Systems;237
7.1;Collaborative Computational Intelligence in Economics;238
7.1.1;Introduction;238
7.1.2;Heterogeneous Agents;239
7.1.2.1;The Three Levels of Collaboration;239
7.1.2.2;Macroscopic Level: Evolving Population;240
7.1.2.3;Microscopic Level: Heterogeneity in Intelligence;250
7.1.2.4;Molecule Level: Hybridization;258
7.1.3;Human and Software Agents;259
7.1.3.1;Mirroring;260
7.1.3.2;Competition;262
7.1.3.3;Collaboration;264
7.1.4;Hybrid Systems;265
7.1.4.1;Nature of Hybridization;265
7.1.4.2;Evolutionary-Based Hybridization;266
7.1.4.3;Semantics-Based Hybrid Systems;269
7.1.4.4;Feature Reduction: Rough GA or GP;270
7.1.5;Concluding Remarks;271
7.1.6;References;272
7.2;IMMUNE: A Collaborating Environment for Complex System Design;279
7.2.1;Introduction;280
7.2.2;Design Planning for Complex Products Development;282
7.2.2.1;The Necessity of Emulating Low Level Product Knowledge for Complex Product Design Planning;284
7.2.2.2;Concurrent Parametric Design;285
7.2.2.3;Simulation Based Engineering, and Complexity Measures;286
7.2.2.4;Deriving a Team Based DSM from a Simulated Parameter Based DSM;289
7.2.3;Radical Innovation;292
7.2.3.1;Holistic Process Monitoring;294
7.2.3.2;Artificial Immune Systems (AIS);297
7.2.4;IMMUNE: A Collaborating Architecture;299
7.2.4.1;Blackboard Architecture;303
7.2.4.2;Control Source;303
7.2.4.3;Agents Structures;306
7.2.5;Implementation and Overall Behavior;309
7.2.5.1;Structuration: Adaptive Organization Structure with Virtual Cross-Functional Teams;310
7.2.5.2;Global Decomposition Strategies and Modality;314
7.2.6;Conclusion;318
7.2.7;References;320
7.3;Bayesian Learning for Cooperation in Multi-Agent Systems;325
7.3.1;Introduction;325
7.3.2;Background;327
7.3.2.1;The Disaster Response Domain;327
7.3.2.2;Single Agent Decision Making;328
7.3.2.3;Coordinated Decision Making;332
7.3.3;Bayesian Learning Models;336
7.3.3.1;Bayesian Learning;336
7.3.3.2;MDPs and POMDPs;337
7.3.4;Bayesian Learning Approximation Using Finite State Machines;341
7.3.4.1;Definitions;341
7.3.4.2;Learning FSMs;342
7.3.4.3;Online Solutions: Best Response;344
7.3.4.4;An Online Learning Algorithm;345
7.3.5;Model Instantiation;347
7.3.6;Experimental Evaluation;349
7.3.6.1;Experimental Setup;350
7.3.6.2;Examining the Learning Rate;350
7.3.6.3;Varying the Sampling Rate;352
7.3.6.4;Varying the Visibility;354
7.3.6.5;Varying the Victim Arrival Rate;356
7.3.6.6;Varying Scaling Factors;356
7.3.7;Conclusions and Future Work;360
7.3.8;References;362
7.4;Collaborative Agents for Complex Problems Solving;365
7.4.1;Introduction;365
7.4.2;Self-interested and Cooperative Multi-Agent Systems;366
7.4.2.1;Traditional Classification;366
7.4.2.2;The Blurred Boundary;367
7.4.2.3;Two Scenarios;367
7.4.3;Collaborative Problem Solving through Agent Cooperation;368
7.4.3.1;Agent Cooperation in Agent Teams: The Scenario;370
7.4.3.2;One-Shot and Long-Term Team-Formation Mechanisms;373
7.4.3.3;Flexible Team-Formation Mechanism;375
7.4.3.4;Experiments;381
7.4.3.5;Summary;384
7.4.4;Collaborative Problem Solving through Agent Competition;384
7.4.4.1;Traditional Agent Negotiation;384
7.4.4.2;Partner Selection in Agent Negotiation;388
7.4.4.3;Behavior Prediction in Agent Negotiation;392
7.4.5;Conclusion;400
7.4.6;References;401
8;Part V Computer Vision;404
8.1;Predicting Trait Impressions of Faces Using Classifier Ensembles;405
8.1.1;Introduction;405
8.1.2;Face Classification: Single Classifier Algorithms and Collaborative Methods;410
8.1.2.1;Single Classifier Algorithms;415
8.1.2.2;Classifier Ensembles;422
8.1.2.3;Resources;424
8.1.3;Study Design;424
8.1.3.1;Step 1: Generation of Stimulus Faces;425
8.1.3.2;Step 2: Assessing Trait Impressions of Stimulus Faces;426
8.1.3.3;Step 3: Division of Stimulus Faces into Trait Class Sets;427
8.1.4;Classification Experiments;429
8.1.4.1;System Architecture;429
8.1.4.2;Results;432
8.1.5;Conclusion;435
8.1.6;Appendix;437
8.1.7;References;438
8.2;The Analysis of Crowd Dynamics: From Observations to Modelling;442
8.2.1;Introduction;442
8.2.2;Background;444
8.2.2.1;Crowd Information Extraction;446
8.2.2.2;Crowd Modelling and Events Inference;449
8.2.2.3;Examples of Bridging the Research;450
8.2.3;Measuring Crowd Motion;451
8.2.3.1;Method 1: Pyramid-Based Interest Points Topological Matching;451
8.2.3.2;Method 2: Using Edge Continuity Constrains of Interest Points;453
8.2.3.3;Comparison of the Two Methods;455
8.2.3.4;Testing Based on Motion Connect Component;457
8.2.4;Modelling Crowd Dynamics;461
8.2.4.1;Statistical Analysis;461
8.2.4.2;Path Discovery;462
8.2.4.3;Self-Organizing Map for Learning Crowd Dynamics;463
8.2.5;Discussion;467
8.2.6;References;468
9;Part VI Communications for CI Systems;474
9.1;Computational Intelligence for the Collaborative Identification of Distributed Systems;475
9.1.1;Introduction;475
9.1.1.1;Sensor Networks: The State of the Art;476
9.1.1.2;Identification of Distributed Systems;478
9.1.2;Modeling a Distributed System by the Karhunen-Lo\'{e}ve Transform;480
9.1.3;Identification of a Distributed System Knowing the Output y;482
9.1.3.1;Neural Network Based Identification;483
9.1.4;Identification of a Distributed System by a Network of Independent Sensors;485
9.1.4.1;Two Sensors (S = 2);486
9.1.4.2;S Sensors;487
9.1.4.3;Best Estimate of the Matrix $\Psi$ Based on the Distributed KLT Algorithm;489
9.1.5;Experimental Results;490
9.1.5.1;First Experiment: Parabolic PDE;490
9.1.5.2;Second Experiment: Hyperbolic PDE;491
9.1.6;Conclusions;498
9.1.7;References;498
9.2;Collaboration at the Basis of Sharing Focused Information: The Opportunistic Networks;501
9.2.1;Introduction;501
9.2.2;The Opportunistic Networks Framework;503
9.2.2.1;A Case Study to Lead the Theory;505
9.2.2.2;An Elementary Mathematical Model;507
9.2.2.3;A Very General Way of Maintaining Memory in a Time Process;509
9.2.2.4;The Timing of the Intentional Process;511
9.2.3;Validating the Model;513
9.2.3.1;Drawing Data and Models from the Literature;513
9.2.3.2;A Homemade Validation;514
9.2.3.3;The Architecture;515
9.2.3.4;Preliminary Matches;516
9.2.3.5;The Statistical Versant;518
9.2.4;Exploiting the Model;520
9.2.5;Conclusions;521
9.2.6;References;522
10;Part VII Artificial Immune Systems;525
10.1;Exploiting Collaborations in the Immune System: The Future of Artificial Immune Systems;526
10.1.1;Introduction;526
10.1.1.1;A Reflection on AIS Today;527
10.1.1.2;Challenges Posed by Real Systems;529
10.1.2;The Natural Immune System;531
10.1.2.1;Innate vs. Adaptive Immunology;532
10.1.2.2;Cooperative Innate Immunology;532
10.1.2.3;Dendritic Cells;534
10.1.3;The Adaptive Immune System: Carneiro’s Networks;536
10.1.3.1;The Cognitive Immune System;541
10.1.4;Interpreting Immune Collaborations in Real-World Applications;544
10.1.4.1;Application of Carneiro’s Model in a Machine Learning Scenario;544
10.1.4.2;The Relationship of the Carneiro Model with Theoretical Machine Learning;546
10.1.5;A Practical Perspective: Application of Innate and Adaptive Immune Mechanisms to WSN;547
10.1.5.1;Immune Approaches to SpeckNets;549
10.1.6;The Future of AIS: Immuno-Engineering;552
10.1.7;Conclusions;553
10.1.8;References;554
11;Part VIII Parallel Evolutionary Algorithms;558
11.1;Evolutionary Computation: Centralized, Parallel or Collaborative;559
11.1.1;Introduction;559
11.1.2;Darwinism - The Unfinished Theory;560
11.1.2.1;The System View of Evolution;563
11.1.3;Evolutionary Algorithms - Centralized, Parallel or Collaborative;565
11.1.4;Co-evolution and Collaboration in Evolution;567
11.1.4.1;Darwin Revisited;567
11.1.4.2;Spatial Population Structures in Evolution Theories;568
11.1.5;The Iterated Prisoner’S Dilemma as an Evolutionary Game;570
11.1.5.1;The Simulation of Spatial Structures Using the Iterated Prisoner’s Dilemma;571
11.1.5.2;The Genetic Representation;572
11.1.5.3;Mathematical Analysis of Structured Populations in Evolutionary Games;573
11.1.5.4;Simulation Results;574
11.1.5.5;The Punctuated Equilibrium Theory;579
11.1.6;Combinatorial Optimization by the PGA;579
11.1.6.1;The Traveling Salesman Problem;580
11.1.6.2;The Graph Partitioning Problem;582
11.1.7;Continuous Function Optimization by Competition;586
11.1.7.1;The BGA for Continuous Parameter Optimization;586
11.1.7.2;Competition between Subpopulations;587
11.1.7.3;The Basic Competition Model of the BGA;588
11.1.7.4;The Extended Competition Model;589
11.1.8;Conclusion;590
11.1.9;References;591
12;Part IX CI for Clustering and Classification;594
12.1;Fuzzy Clustering of Likelihood Curves for Finding Interesting Patterns in Expression Profiles;595
12.1.1;Introduction;595
12.1.2;Background to Quantitative Proteomics and iTRAQ$^{TM}$ Based Likelihood Curves;597
12.1.2.1;Proteomics;598
12.1.2.2;Identification and Characterisation of Proteins Based on LC-MS;598
12.1.2.3;Quantification of Peptides and Proteins;599
12.1.2.4;Impreciseness of Regulatory Information: Intensity-Dependent Noise;602
12.1.2.5;Calculation and Visualisation of Regulatory Information;605
12.1.3;Fuzzy Cluster Analysis;608
12.1.4;Fuzzy Clustering of Likelihood Curves;610
12.1.4.1;Generation of Prototypes;611
12.1.4.2;Validity Measures;613
12.1.4.3;Examples;614
12.1.5;Conclusions;617
12.1.6;References;618
12.2;A Hybrid Rule-Induction/Likelihood-Ratio Based Approach for Predicting Protein-Protein Interactions;619
12.2.1;Introduction;619
12.2.1.1;Computational Prediction of Protein-Protein Interactions;620
12.2.1.2;Overview of the Proposed Method;621
12.2.1.3;Organisation;622
12.2.2;Classification Rule Discovery Algorithms;622
12.2.2.1;Separate-and-Conquer Approach;623
12.2.2.2;Divide-and-Conquer Approach;623
12.2.3;Protein Interaction Data and Predictive Features;625
12.2.4;A New Hybrid Rule Induction/Likelihood-Ratio Based Method;626
12.2.4.1;From Naive Bayes to a Likelihood Based Approach for the Prediction of Protein-Protein Interactions;626
12.2.4.2;Generating Classification Rules for Protein-Protein Interaction Prediction;627
12.2.4.3;Classification Rule Discovery as a Binning Method for a Likelihood-based Approach;628
12.2.5;Results and Discussion;628
12.2.6;Conclusions;630
12.2.7;References;631
12.3;Improvements in Flock-Based Collaborative Clustering Algorithms;634
12.3.1;Introduction;634
12.3.2;Swarm Intelligence Clustering;636
12.3.2.1;Particle Swarm Clustering;637
12.3.2.2;Ant Clustering;638
12.3.3;Flocks of Agents for Data Visualization and Clustering;640
12.3.3.1;Flocks of Agents Based-Data Visualization;641
12.3.3.2;Flocks of Agents-Based Clustering;641
12.3.4;Improved Distance Threshold Estimates;646
12.3.4.1;Alternative Fixed Thresholding;646
12.3.4.2;Adaptive Thresholding Using FClust-Annealing;646
12.3.5;The (K-means/FClust) Hybrid Algorithm;647
12.3.5.1;K-Means Algorithm;647
12.3.5.2;(K-means+FClust) Hybrid;648
12.3.5.3;Stopping Criterion;649
12.3.6;Experimental Results;650
12.3.6.1;Datasets;650
12.3.6.2;Post Processing;651
12.3.6.3;Results;652
12.3.7;Conclusions and Future Work;664
12.3.8;References;665
12.4;Combining Statistics and Case-Based Reasoning for Medical Research;668
12.4.1;Introduction;668
12.4.1.1;Case-Based Reasoning in Medicine;669
12.4.1.2;The ISOR Approach;670
12.4.2;Incremental Development of an Explanation Model for Exceptional Dialysis Patients;672
12.4.2.1;Setting up a Model;673
12.4.2.2;Setting up a Case Base;675
12.4.2.3;Another Problem;677
12.4.2.4;Example;678
12.4.3;Illustration of ISOR’s Program Flow;679
12.4.4;Missing Data;681
12.4.4.1;The Data Set;682
12.4.4.2;Restoration of Missing Data;682
12.4.5;Results;686
12.4.5.1;Experimental Results;686
12.4.5.2;Restoration of Real Missing Data and Setting up a New Model;688
12.4.6;Conclusion;689
12.4.7;References;690
12.5;Collaborative and Experience-Consistent Schemes of System Modelling in Computational Intelligence;692
12.5.1;Introductory Comments;693
12.5.2;Collaborative Clustering;695
12.5.3;The General Flow of Collaborative Processing;698
12.5.4;Algorithmic Aspects of Collaborative Clustering;699
12.5.4.1;The Computing Scheme;699
12.5.4.2;Evaluation of the Quality of Collaboration: Striking a Sound Compromise between Global and Local Characteristics of Data;700
12.5.4.3;Fuzzy Sets of Type-2 in the Quantification of the Effect of Collaboration;702
12.5.4.4;Collaborative Clustering in Presence of Different Levels of Information Granularity;703
12.5.5;Hierarchical Clusters of Clusters;705
12.5.6;Experience-Consistent Fuzzy Modeling;706
12.5.6.1;The Consistency-Based Optimization of Local Regression Models;708
12.5.6.2;The Alignment of Information Granules;712
12.5.6.3;Characterization of Experience-Consistent Models through its Granular Parameters;712
12.5.7;Experience-Consistent Design of Radial-Basis Function Neural Networks;714
12.5.8;Conclusions;717
12.5.9;References;717
13;Index;719