Adaptive Intelligent Systems (eBook)

Front Cover 1
Adaptive Intelligent Systems 4
Copyright Page 5
Table of Contents 8
PREFACE 6
Chapter 1. Artificial Neural Networks and ARIMA-Models within the Field of Stock Market Prediction - A Comparison 10
Abstract: 10
1. CHARACTERIZATION OF THE STOCK MARKET PREDICTION 10
2. DATA MATERIAL AND METHODS FOR PROGNOSIS 11
3. STOCK PREDICTION WITH ANN 12
4. STOCK PREDICTION WITH ARIMA-MODELS 19
5. CONCLUSIONS 25
REFERENCES 26
Chapter 2. A Decision Support System Building Tool with Fuzzy Logic and Its Application to Chart Technical Analysis 28
Abstract 28
1. Introduction 28
2. An Intelligent Decision Support System with Fuzzy Logic for Financial Fields 29
3. A Decision Support System Building Tool with Fuzzy Logic 31
4. Application to Chart Technical Analysis 37
5. Conclusions 40
References 40
Section 1: Debate I: That Neural Networks are an Applicable Technology Today 42
Chapter 3. Man-Machine Synergy in Action 52
Abstract 52
1. MOTIVATION 52
2. OUR APPROACH 53
3. SAMPLE APPLICATION 60
4. LESSONS 63
5. CONCLUSION 65
References 66
Chapter 4. KNOWLEDGE STRUCTURING AND FORMALIZATION IN A DISTRIBUTED ENVIRONMENT: An application to the firms results rating 68
Abstract 68
1. INTRODUCTION 68
2. GENERAL PRESENTATION OF THE METHOD 69
3. DETAILED PRESENTATION OF THE METHOD 70
4. AN EXAMPLE OF IMPLEMENTATION : SACRE 76
5. CONCLUSION 80
REFERENCES 81
Chapter 6. An Adaptable Reporting Architecture 84
Abstract 84
1. Introduction 84
2. General Reporting System 86
3. The Next Generation FRA 90
4. Conclusions 94
Chapter 7. FIRCO: The Intelligent Transformation of D Fields Into A Fields 96
Abstract 96
1. QUALITY CONTROL APPLIED TO SWIFT MESSAGE TRANSFERS 96
2. FIRCO: TRANSFORMING D FIELDS 98
3. QUALITY METRICS 104
4. CONCLUSIONS 105
5. REFERENCES 105
6. ACKNOWLEDGEMENTS 106
Section 2: Debate II. That Classical Software Engineering Methods Are Suitable for Developing Adaptive Systems 108
Chapter 8. The Schematic Programming Tool: An Application of A.I. to Software Engineering 120
Abstract 120
1. INTRODUCTION 120
2. AN OVERVIEW 121
3. THE MAIN CONCEPTS OF SPT 122
4. AN EXAMPLE 125
5. CONCLUSIONS 126
6. REFERENCES 128
Chapter 9. Neural Network Futures Trading - A Feasibility Study 130
ABSTRACT 130
1. BACKGROUND 131
2. ENVIRONMENT OF THE PROJECT DOMAIN 131
3. REASONS FOR CHOOSING A NEURAL NETWORK 132
4. BACK PROPAGATION LEARNING 132
5. SYSTEM CRITERIA & NETWORK SPECIFICATIONS
6. TRADING STRATEGY EMPLOYED AND ITS PLACE IN THE REAL ENVIRONMENT 135
7. DATA ITEMS CHOSEN 136
8. DATA MANIPULATION 137
9. METHOD EMPLOYED 138
10. EVALUATION CRITERIA 139
11. EVALUATION OF THE STUDY 141
12. REFERENCES 141
Chapter 10. An Application of the AQ Machine Learning Methodology on the Stock Market 142
Abstract 142
1. INTRODUCTION 142
2. PREDICTING THE STOCK MARKET 143
3. MACHINE LEARNING 146
4. CASES 150
5. APPLYING AQ AND LEARNING RESULTS 153
6. CONCLUSION 154
Acknowledgements 155
REFERENCES 155
Chapter 11. Prediction of Stock Market Index Changes 158
Abstract 158
1. Introduction 158
2. Prediction of Stock Market Index 159
3. Instance-Based Learning 160
4. Nested Generalized Exemplars 162
5. Artificial Neural Networks 165
6. Conclusion 166
References 168
Appendix: Generated Rules 169
Chapter 12. Credit risk and lending in an artificial adaptive banking system 170
Abstract 170
1. INTRODUCTION 170
2. THE BANKING SYSTEM 172
3. BACKGROUND ON NEURAL NETWORKS 174
4. THE LEARNING OF THE BANKS 177
5. THE PARAMETERS OF THE SYSTEM 178
6. RESULTS 179
7. REFERENCES 184
APPENDIX 185
Chapter 13. Consumer Loan Analysis Using Neural Networks 186
1. Introduction 186
2. Neural Networks 187
3. A Survey of Techniques for Consumer Loan Analysis 189
4. Empirical Comparison of Different Approaches 191
5. Empirical Results for Different Configurations 192
6. A Possible System Configuration 194
7. Concluding Remarks 197
8. Addendum 197
References 197
Appendix 199
Chapter 14. Improving The Neural Network Testing Process 202
Abstract 202
1. HOW IMPORTANT IS INDUCTION FOR BANKING ESTABLISHMENTS ? 202
2. HOW CAN INDUCTION TECHNIQUES BE COMPARED ? 203
3. METHODOLOGY ASSETS 204
4. THE BACKPROPAGATION PROCESS 204
5. WHY IS DIFFICULT TO CONDUCT EXPERIMENTS ? 205
6. PARAMETERS TO SET 206
7. WHAT ARE THE POSSIBLE ANSWERS? 208
8. OBSERVING THE NETWORK 209
9. DEBUGGING A NETWORK 210
10. CASE STUDY 212
11. CONCLUSION 213
12. REFERENCES 213
Chapter 15. DeTerminator: a Decision Support System and Tool-Kit using the ProFuSE method 216
Abstract 216
1. Introduction 216
2. Design of DeTerminator 217
3. Implementation of DeTerminator 224
4. Notes on Adaptivity 225
5. Empirical Test of Modelling Tool 226
6. Conclusion and Outlook 235
7. Acknowledgments 237
8. Bibliography 237
Section 3: Debate III: That Rule-Based Systems are an Evolutionary Dead End in the Development of Intelligent Systems 238
Chapter 16. AN EXPERT SYSTEM FOR PERSONAL FINANCIAL ASSET MANAGEMENT USING ANALOGICAL, QUALITATIVE AND CAUSAL REASONING 248
Abstract 248
1. INTRODUCTION 248
2. MANAGEMENT OF PERSONAL FINANCIAL ASSETS 249
3. STRUCTURE OF THE SYSTEM 250
4. MODELS OF CLIENTS 251
5. MODELS OF FINANCIAL ITEM 254
6. BEHAVIOUR OF THE SYSTEM 254
7. CONCLUSION 257
8. Acknowledgements 257
9. REFERENCES 258