Adaptive Control Approach for Software Quality Improvement (Series on Software Engineering & Knowledge Engineering)

CONTENTS......Page 8
Preface......Page 6
1. Introduction and Related Work......Page 10
2.1. Event Sequence Graphs......Page 13
2.2. Neural Network-Based Clustering......Page 14
3. Competitive Learning......Page 16
3.1. Distance-Based Competitive Learning Algorithm......Page 18
3.2. Angle-Based Competitive Learning Algorithm......Page 19
3.3. Adaptive Competitive Learning......Page 20
Adaptive Competitive Learning Algorithm......Page 21
4. Prioritized ESG-Based Testing......Page 22
4.2. Definition of Importance Degree and Preference......Page 23
5. A Case Study......Page 24
5.1. Derivation of the Test Cases......Page 25
5.2. Determination of Attributes of Events......Page 26
5.4. Indirect Determination of Preference Degrees......Page 27
6. Conclusions and Future Work......Page 28
References......Page 29
1. Introduction......Page 32
2.1. Static V&V Approaches......Page 37
2.2. Dynamic V&V Approaches......Page 38
2.3. V&V of Neural Networks......Page 40
3.1. Neural Network-Based Flight Control......Page 41
3.2.1. Dynamic Cell Structure Network......Page 43
3.3. Failure Detection Using Support Vector Data Description......Page 46
3.4. Evaluating Network’s Learning Performance......Page 54
3.4.1. A Sensitivity Metric for DCS Networks......Page 55
3.4.2. A Sensitivity Metric for Sigma-Pi Networks......Page 56
3.5.1. Validity Index for DCS Networks......Page 57
3.5.2. Bayesian Confidence Tool for Sigma-Pi Networks......Page 58
4. Conclusions......Page 62
References......Page 63
1. Introduction......Page 66
2. Adaptive Random Testing......Page 70
2.1. Distance-Based Adaptive Random Testing......Page 71
2.2. Restriction-Based Adaptive Random Testing......Page 72
2.3. Overheads......Page 74
2.4. Filtering......Page 75
2.5. Forgetting......Page 77
2.6. Mirror ART......Page 78
2.7. Probabilistic ART......Page 79
2.8. Fuzzy ART......Page 80
Acknowledgements......Page 81
References......Page 82
1. Introduction......Page 86
2. Basic Elements......Page 88
3. General Approach......Page 90
4. Middleware-Based Transparent Shaping......Page 92
4.1. ACT Architectural Overview......Page 93
4.2. ACT Core Components......Page 94
Proxies......Page 95
4.3. ACT Operation......Page 96
4.4. ACT/J Implementation......Page 97
4.5. ACT/J Case Study......Page 98
5.1. TRAP/J Architectural Overview......Page 102
5.2. TRAP/J Run-Time Model......Page 104
5.3. TRAP/J Case Study......Page 107
Making ASA Adapt-Ready......Page 108
Generated Aspect......Page 109
Generated Wrapper-Level Class......Page 110
Generated Metalevel Class......Page 111
Balancing QoS and Energy Consumption......Page 113
6. Discussion......Page 115
Acknowledgements......Page 118
References......Page 119
1. Neural Network Rule Extraction......Page 124
1.1. Background on Rule Extraction......Page 125
2. Rule Extraction for System Verification and Validation......Page 127
2.1. An Example of Rule Extraction for the Dynamic Cell Structure Neural Network Used in a System......Page 128
2.1.1. Refining the Algorithm......Page 129
3. Applying Rule Extraction in a Tool for Verification and Validation......Page 133
3.1. Describing a Neural Network with Metadata Expressions......Page 134
3.2. Building a Tool for Rule Extraction......Page 135
3.3.2. Extract Rules......Page 137
3.3.3. Analyze Rules......Page 139
Scenario 1: Human Understandable Rules Led to Identi.cation of Coding Error.......Page 140
Scenario 2: Machine Understandable Rules Led to Identi.cation of Two Coding Errors.......Page 141
5.2. Health and Status Monitoring of the Neural Network......Page 144
5.3. Extracted Rules as Basis for Expert Systems......Page 145
6. Conclusion......Page 146
References......Page 148
Appendix A......Page 150
6. Requirements Engineering Via Lyapunov Analysis for Adaptive Flight Control Systems Giampiero Campa, Marco Mammarella, Mario L. Fravolini and Bojan Cukic......Page 154
1. Introduction......Page 155
2.1. The Plant, the Closed Loop System and the Error Dynamics......Page 156
2.2. The Linear Controller, the Closed Loop System and the Error Dynamics......Page 157
2.3. The Uncertainty......Page 158
2.5. The Adaptive Augmentation......Page 159
3. The Lyapunov Analysis......Page 160
3.1. Typical “Completion of Squares” Bounds Formulation and its Limitations......Page 161
3.2. A Better Characterization of the Return Set......Page 163
3.3. Boundedness Conditions......Page 164
3.3.1. Extreme Points of the Boundary and Semi-Axes of the Ellipsoid......Page 165
4. Case Study......Page 166
4.0.1. 2D Bounds Calculation and Visualization......Page 169
5. Conclusions......Page 172
Books......Page 173
Computer Software......Page 174
1. Introduction......Page 176
1.2. Related Work......Page 179
2. General Modeling Strategy......Page 180
2.1. Mathematical Modeling of Productive Capability......Page 185
2.2. State Model of Productive Capability......Page 188
2.3. State Model of a Queue......Page 189
2.4. Normalization of Work Items......Page 191
2.5. Managing Dependencies and Scheduling Constraints within the Model......Page 192
2.6. An Algebraic Model of Activity Coordination......Page 194
2.7. Defect Modeling and the Failure Analysis Activity......Page 196
2.8. An Algebraic Model for the Example Defect Population Estimation Component......Page 200
2.9. An Algebraic Model of the Defect Detection Component......Page 201
3. Assembling the Model......Page 202
4.1. Simulation Method......Page 203
4.2. Simulation Results......Page 205
4.2.1. Feature Coding......Page 206
4.2.2. Test Case Coding......Page 208
4.2.3. New Test Case Execution......Page 210
4.2.4. Regression Test Case Execution......Page 216
4.2.5. Defect Introduction, Defect Detection, and Failure Analysis......Page 220
4.2.6. Feature Correction......Page 225
4.2.7. Test Case Correction......Page 229
5. Model Calibration......Page 230
5.1. Calibrating with Ratio Scale Data......Page 231
5.2. Calibrating with Interval Scale Data......Page 232
6. Discussion......Page 235
References......Page 236
Appendix A: Modeling the Example Process......Page 239
Appendix B: Simulation Study Parameters......Page 244
1. Introduction......Page 248
2. Current State of the Art......Page 251
3.1. Workflow Virtual Machine......Page 252
4. Synthesizing Software Modules for Proactive Monitoring and Control of Workflow Execution in ASBS......Page 255
4.1. Workflow Execution, Monitoring and Control in ASBS......Page 257
4.2. Synthesizing WF Monitors......Page 265
4.3. Synthesizing WF Controllers......Page 270
5. Conclusions and Future Work......Page 272
References......Page 273
1. Introduction......Page 276
Classical Software Failure Mechanics......Page 278
Fundamentals of Software Aging......Page 281
3. Accelerated Life Tests......Page 286
Accelerated Degradation Tests......Page 291
4. Case Study......Page 297
Numerical Results......Page 300
5. Final Remarks......Page 305
References......Page 306