Fuzzy Engineering Expert Systems with Neural Network Applications

FUZZY ENGINEERING EXPERT SYSTEMS WITH NEURAL NETWORK APPLICATIONS......Page 3
CONTENTS......Page 9
Preface......Page 17
Acknowledgments......Page 19
1.1 Origin of Artificial Intelligence......Page 23
1.2 Human Intellligence versus Machine Intelligence......Page 25
1.3 The First AI Conference......Page 27
1.4 Evolution of Smart Programs......Page 28
1.5 Branches of Artificial Intelligence......Page 30
1.6 Neural Networks......Page 31
1.7 Emergence of Expert Systems......Page 32
1.7.1 Embedded Expert Systems......Page 34
2.1 Expert Systems Process......Page 35
2.2 Expert Systems Characteristics......Page 36
2.2.1 Domain Specificity......Page 37
2.3 Expert Systems Structure......Page 38
2.3.2 Benefits of Expert Systems......Page 41
2.3.3 Transition from Data Processing to Knowledge Processing......Page 42
2.4 Heuristic Reasoning......Page 43
2.5 User Interface......Page 44
2.5.4 Application Roadmap......Page 45
2.5.5 Symbolic Processing......Page 46
3.1 Problem Identification......Page 50
3.2 Problem Analysis......Page 51
3.2.4 Frequency of Problem Occurrence......Page 52
3.2.6 Data and Knowledge Availability......Page 53
3.3 Data Requirement Analysis......Page 56
3.4 Expert System Justification......Page 58
3.4.1 Problem-Selection Guidelines......Page 69
4.1.1 The Knowledge Engineer......Page 71
4.1.2 Knowledge Characteristics......Page 72
4.1.3 Choosing the Expert......Page 73
4.1.4 Knowledge Extraction versus Knowledge Acquisition......Page 75
4.2.1 Interviews......Page 76
4.2.2 Open-Ended Interviews......Page 77
4.2.4 Task Performance and Protocols......Page 78
4.2.6 Constrained Task......Page 79
4.2.9 Documentation and Analysis of Acquired Knowledge......Page 80
4.3 Knowledge-Acquisition Meetings......Page 81
4.4 Group Knowledge Acquisition......Page 83
4.4.1 Brainstorming......Page 84
4.4.2 Delphi Method......Page 85
4.4.3 Nominal Group Technique......Page 86
4.5.1 Knowledge Elicitation Tools......Page 88
4.5.2 Induction-by-Example Tools......Page 89
4.6.1 Types of Knowledge......Page 90
4.7 Knowledge-Representation Models......Page 91
4.7.1 Semantic Networks......Page 92
4.7.2 Frames......Page 93
4.7.3 Scripts......Page 95
4.7.4 Rules......Page 96
4.7.5 Predicate Logic......Page 98
4.7.6 O-A-V Triplets......Page 99
4.7.8 Specialized Representation Techniques......Page 100
4.8 Concept of Knowledge Sets......Page 109
4.8.1 Properties of Knowledge Sets......Page 110
4.9 Reasoning Models......Page 119
5.1 Human Reasoning and Probability......Page 125
5.2 Bayesian Approach to Handling Uncertainty......Page 126
5.3 Decision Tables and Trees......Page 128
5.4 Dempster–Shafer Theory......Page 135
5.4.3 Uncertainty of A......Page 138
5.4.6 Doubt Function......Page 139
5.5 Certainty Factors......Page 141
5.6 Fuzzy Logic......Page 144
5.6.1 Definition of Fuzzy Set......Page 146
6.1.1 Crisp Sets......Page 155
6.1.2 Fuzzy Sets......Page 158
6.1.3 Fuzzy Set Construction......Page 162
6.1.4 Fuzzy Set Operations......Page 163
6.1.5 a-Cuts......Page 165
6.1.6 Extension Principle......Page 167
6.2 Fuzzy Operations......Page 168
6.2.1 Fuzzy Complement......Page 169
6.2.2 Fuzzy Intersection......Page 170
6.2.3 Fuzzy Union......Page 173
6.2.4 Duality......Page 175
6.2.5 Fuzzy Implication......Page 176
6.2.6 Fuzzy Aggregation......Page 178
6.3.1 Fuzzy Number Representation......Page 179
6.3.2 Fuzzy Number Operation......Page 180
6.4.1 Definition of a Fuzzy Relation......Page 183
6.4.2 Binary Relation......Page 184
6.4.3 Operations with Relations......Page 186
6.5.1 Believability and Plausibility......Page 187
6.5.3 Fuzziness......Page 190
6.5.4 Nonspecificity......Page 192
6.6 Fuzzy Logic......Page 193
6.6.2 Unconditional Fuzzy Propositions......Page 194
6.6.3 Conditional Fuzzy Propositions......Page 195
6.6.4 Selection of Implication Operator......Page 198
6.6.5 Multiconditional Reasoning......Page 199
6.7 Summary......Page 201
7.1 Introduction......Page 202
7.1.1 Definition of a Neurode......Page 203
7.1.2 Variations of a Neurode......Page 204
7.2.2 McCulloch–Pitts Neurodes as Boolean Components......Page 205
7.2.3 Single Neurode as Binary Classifier......Page 206
7.2.4 Single-Neurode Perceptron......Page 208
7.3.2 Associative Memory......Page 209
7.3.3 Correlation Matrix Memory......Page 211
7.3.5 Widrow–Hoff Approach......Page 212
7.3.7 Adaptive Correlation Matrix Memory......Page 213
7.4 Self-Organizing Networks......Page 214
7.4.1 Principal Components......Page 215
7.4.2 Clustering by Hebbian Learning......Page 216
7.4.3 Clustering by Oja’s Normalization......Page 217
7.4.4 Competitive Learning Network......Page 219
7.5.1 Multiple-Layer Perceptron......Page 221
7.5.2 XOR Example......Page 222
7.5.3 Back-Error Propagation......Page 223
7.5.4 Variations in the Back-Error Propagation Algorithm......Page 224
7.5.5 Learning Rate and Momentum......Page 225
7.5.7 Counterpropagation Network......Page 227
7.6 Radial Basis Networks......Page 228
7.6.1 Interpolation......Page 229
7.6.2 Radial Basis Network......Page 230
7.7 Single-Layer Feedback Network......Page 233
7.7.1 Single-Layer Feedback Network......Page 234
7.7.2 A Discrete Single-Layer Feedback Network......Page 235
7.7.3 Bidirectional Associative Memory......Page 237
7.7.4 Hopfield Network......Page 238
7.8 Summary......Page 241
7.9 References......Page 242
8.1 Technology Comparisons......Page 243
8.2 Neurons Performing Fuzzy Operations......Page 245
8.2.1 Neurons Emulating Fuzzy Operations......Page 246
8.2.2 Neurons Performing Fuzzy Operations......Page 248
8.3.1 Regular Neural Network with Crisp Input and Output......Page 249
8.3.2 Regular Neural Network with Fuzzy Input and Output......Page 250
8.3.3 Fuzzy Inference Network......Page 251
8.3.4 ANFIS......Page 252
8.3.5 Applications......Page 253
8.4 Clustering and Classification......Page 254
8.4.1 Classification......Page 255
8.4.2 Multilayer Fuzzy Perceptron......Page 257
9.1 Introduction......Page 259
9.2 Binary Genetic Algorithm......Page 261
9.2.1 Genetic Representation......Page 263
9.2.3 Fitness Check and Cost Evaluation......Page 264
9.2.4 Mating Pool......Page 265
9.2.5 Pairing......Page 266
9.2.6 Mating......Page 268
9.2.7 Mutation......Page 270
9.2.8 The Next Generation......Page 271
9.2.9 Performance......Page 272
9.2.10 Enhancements......Page 273
9.3 Continuous Genetic Algorithm......Page 274
9.3.1 Genetic Representation......Page 275
9.3.2 Mating......Page 276
9.3.4 Performance......Page 278
9.3.5 Enhancements......Page 279
9.4.1 Evolutionary Strategies......Page 281
9.4.2 Evolutionary Programming......Page 282
9.5 Summary......Page 285
10.1 Introduction......Page 286
10.2 Model Description......Page 287
10.3 Evolutionary Algorithm......Page 289
10.4 Process Simulation......Page 291
10.5 Fuzzy Logic Evaluation......Page 293
11.1 Introduction......Page 296
11.2.2 Fitness Evaluation......Page 298
11.2.3 Reproduction and Generation Evolution......Page 299
11.3 Experiments and Results......Page 300
11.4 Conclusion......Page 303
References......Page 304
Index......Page 311