Knowledge Mining Using Intelligent Agents

CONTENTS......Page 10
PREFACE......Page 6
1.1. Knowledge and Agent......Page 12
1.2. Knowledge Mining from Databases......Page 13
1.2.1. KMD tasks......Page 15
1.2.1.2. Classification......Page 16
1.2.1.6. Regression......Page 17
1.3.1. Evolutionary computing......Page 18
1.3.2. Swarm intelligence......Page 19
1.3.2.1. Particle Swarm Optimization......Page 22
1.3.2.2. Ant Colony Optimization (ACO)......Page 23
1.3.2.3. Artificial Bee Colony (ABC)......Page 26
1.3.2.4. Artificial Wasp Colony (AWC)......Page 29
1.3.2.5. Artificial Termite Colony (ATC)......Page 30
1.4. Summary......Page 33
References......Page 34
2.1. Introduction......Page 38
2.2.1. Knowledge discovery and data mining......Page 40
2.2.2. Evolutionary computation......Page 41
2.2.3. Intrusion detection systems......Page 44
2.3.1. Feature selection......Page 45
2.3.2. Classification......Page 46
2.3.2.1. Representation......Page 47
2.3.2.2. Learning approaches......Page 49
2.3.2.3. Rule discovery......Page 50
2.3.3. Regression......Page 53
2.3.4. Clustering......Page 55
2.3.5. Comparison between classification and regression......Page 56
2.4.1. Evolutionary operators......Page 57
2.4.2. Niching......Page 59
2.5. Fitness Function......Page 60
Acknowledgment......Page 62
References......Page 63
3.1. Introduction......Page 72
3.2. Evolving Neural Network......Page 74
3.2.1. The evolution of connection weights......Page 76
3.2.2. The evolution of architecture......Page 77
3.2.4. Evolution of learning rules......Page 79
3.2.5. Evolution of algorithmic parameters......Page 80
3.3. Evolving Neural Network using Swarm Intelligence......Page 81
3.3.1. Particle swarm optimization......Page 82
3.3.2. Swarm intelligence for evolution of neural network architecture......Page 83
3.3.2.2. Fitness evaluation......Page 84
3.3.3. Simulation and results......Page 85
3.4. Evolving Polynomial Network (EPN) using Swarm Intelligence......Page 89
3.4.1. GMDH-type polynomial neural network model......Page 91
3.4.2. Evolving polynomial network (EPN) using PSO......Page 92
3.4.3. Parameters of evolving polynomial network (EPN)......Page 94
3.4.3.1. Highest degree of the polynomials......Page 95
3.4.3.2. Number of terms in the polynomials......Page 96
3.4.4. Experimental studies for EPN......Page 97
3.5. Summary and Conclusions......Page 101
References......Page 102
4.1. Introduction......Page 110
4.2. The Alloy Optimal Design Problem......Page 112
4.3. Neurofuzzy Modeling for Mechanical Property Prediction......Page 113
4.3.1. General scheme of neurofuzzy models......Page 114
4.3.2. Incorporating knowledge into neurofuzzy models......Page 117
4.3.3. Property prediction of alloy steels using neurofuzzy models......Page 118
4.3.3.1. Tensile strength prediction for heat-treated alloy steels......Page 119
4.3.3.2. Impact toughness prediction for heat-treated alloy steels......Page 120
4.4. Introduction to Multi-Objective Optimization......Page 125
4.5.1. Particle swarm optimization algorithm......Page 126
4.5.2. Adaptive evolutionary particle swarm optimization (AEPSO) algorithm......Page 128
4.5.3. Comparing AEPSO with some leading multi-objective optimization algorithms......Page 131
4.6. Multi-Objective Optimal Alloy Design Using AEPSO......Page 134
4.6.1. Impact toughness oriented optimal design......Page 135
4.6.2. Optimal alloy design with both tensile strength and impact toughness......Page 136
4.7. Conclusions......Page 138
References......Page 139
5.1. Introduction......Page 144
5.2. Related Research......Page 146
5.3.1. Naive Bayesian classifier......Page 150
5.3.2. Intrusion detection system......Page 152
5.3.2.1. Architecture of IDS......Page 153
5.3.2.3. Effectiveness......Page 154
5.3.3. Feature selection......Page 155
5.3.4. Particle swarm optimization......Page 156
5.4.1. Adaptive PSO......Page 158
5.4.2. Hybrid APSO......Page 159
5.4.3. Learnable Bayesian classifier in IDS......Page 161
5.5.1. Description of intrusion data......Page 162
5.5.1.3. User to root attacks......Page 163
5.5.2. System parameters......Page 164
5.5.3. Results......Page 166
5.6. Conclusions and Future Research Directions......Page 167
References......Page 168
6.1. Introduction......Page 172
6.2. Mining Knowledge Using Data Mining Techniques......Page 174
6.3. Association Rule Mining......Page 176
6.4. Measuring Interestingness......Page 182
6.5. Classification......Page 184
6.6. Ensemble of Classifier......Page 191
6.7. Clustering......Page 199
Types of Clustering Algorithms:......Page 200
Algorithm description:......Page 201
EM (Expectation Maximization) Clustering......Page 204
References......Page 208
7.1. Introduction......Page 212
7.2.1. State variables......Page 213
7.2.3. Constraints......Page 215
7.3. Particle Swarm Optimization......Page 217
7.3.2. Evolutionary PSO EPSO......Page 218
7.3.3. Adaptive PSO(APSO)......Page 219
7.4. Optimal Operational Planning for Energy Plants Using PSO......Page 220
7.5.1. Simulation conditions......Page 221
7.5.2. Simulation results......Page 222
7.6. FeTOP — Energy Management System......Page 223
7.7. Conclusions......Page 225
References......Page 226
8.1. Introduction......Page 228
8.1.1. Definition......Page 231
8.2.1. Enumerative algorithms......Page 234
8.2.3. Sampling......Page 235
8.2.4. Feature selection based on information theory......Page 238
8.2.5. Floating search for feature selection......Page 240
8.2.8. Feature selection with dynamic mutual information......Page 241
8.2.9. Learning to classify by ongoing feature selection......Page 242
8.2.10. Multiclass MTS for simultaneous feature selection and classification......Page 243
8.3.1. Genetic algorithm for feature selection......Page 244
8.3.2. ELSA......Page 245
8.3.3. Neural network for feature selection......Page 246
8.4. Hybrid Algorithm for Feature Selection......Page 247
8.4.1. Neuro-Fuzzy feature selection......Page 248
8.5. Multi-Objective Genetic Algorithm for Feature Selection......Page 250
8.6. Parallel Genetic Algorithm for Feature Selection......Page 252
Self-adaptive genetic algorithm for clustering (SAGA).......Page 256
8.7. Unsupervised Techniques for Feature Selection......Page 257
8.8. Evaluation functions......Page 259
8.9. Summary and Conclusions......Page 261
References......Page 262
9.1. Introduction......Page 270
9.2. Fuzzy Net Architecture......Page 273
9.3. Particle Swarm Optimization......Page 277
9.3.2. Binary particle swarms......Page 279
9.3.3. Hybrids and adaptive particle swarms......Page 280
9.3.4. PSOs with diversity control......Page 282
9.4. Fuzzy Swarm Net Classifier......Page 283
9.5. Polynomial Neural Network......Page 286
9.6. Classification with Optimized Polynomial Neural Fuzzy Swarm Net......Page 288
9.7.1. Description of the datasets......Page 291
9.8. Conclusion......Page 296
References......Page 297
10.1. Introduction......Page 308
10.2. Overview of Test Case Design......Page 310
10.2.1. Path wise test data generators......Page 311
10.3.1. Introduction to genetic algorithms......Page 314
10.3.2. Overview of genetic algorithms......Page 315
10.4. Path Wise Test Data Generation Based on GA......Page 317
10.5. Summary......Page 323
References......Page 324