Domain Driven Data Mining (eBook)

Preface 6
Acknowledgements 8
Contents 9
Challenges and Trends 15
1.1 Introduction 15
1.2 KDD Evolution 16
1.3 Challenges and Issues 17
1.4 KDD Paradigm Shift 25
1.5 Towards Domain Driven Data Mining 31
1.6 Summary 39
D3M Methodology 40
2.1 Introduction 40
2.2 D3M Methodology Concept Map 40
2.3 D3M Key Components 41
2.4 D3M Methodological Framework 53
2.5 Summary 60
Ubiquitous Intelligence 61
3.1 Introduction 61
3.2 Data Intelligence 61
3.3 Domain Intelligence 67
3.4 Network Intelligence 71
3.5 Human Intelligence 74
3.6 Organizational Intelligence 77
3.7 Social Intelligence 79
3.8 Involving Ubiquitous Intelligence 81
3.9 Summary 84
Knowledge Actionability 86
4.1 Introduction 86
4.2 Why Knowledge Actionability 87
4.3 RelatedWork 88
4.4 Knowledge Actionability Framework 89
4.5 Aggregating Technical and Business Interestingness 98
4.6 Summary 101
D3M AKD Frameworks 103
5.1 Introduction 103
5.2 Why AKD Frameworks 104
5.3 RelatedWork 106
5.4 A System View of Actionable Knowledge Discovery 107
5.5 Actionable Knowledge Discovery Frameworks 111
5.6 Case Studies 119
5.7 Discussions 120
5.8 Summary 122
Combined Mining 123
6.1 Introduction 123
6.2 Why Combined Mining 124
6.3 Problem Statement 127
6.4 The Concept of Combined Mining 131
6.5 Multi-Feature Combined Mining 136
6.6 Multi-Method Combined Mining 142
6.7 Case Study: Mining Combined Patterns in E-Government Service Data 149
6.8 RelatedWork 149
6.9 Summary 152
Agent-Driven Data Mining 154
7.1 Introduction 154
7.2 Complementation between Agents and Data Mining 154
7.3 The Field of Agent Mining 156
7.4 Why Agent-Driven Data Mining 159
7.5 What Can Agents Do for Data Mining? 161
7.6 Agent-Driven Distributed Data Mining 163
7.7 Research Issues in Agent Driven Data Mining 168
7.8 Case Study 1: F-Trade – An Agent-Mining Symbiont for Financial Services 169
7.9 Case Study 2: Agent-based Multi-source Data Mining 170
7.10 Case Study 3: Agent-based Adaptive Behavior Pattern Mining by HMM 171
7.11 Research Resources on Agent Mining 176
7.12 Summary 177
Post Mining 179
8.1 Introduction 179
8.2 Interestingness Measures 180
8.3 Filtering and Pruning 182
8.4 Visualisation 184
8.5 Summarization and Representation 185
8.6 Post-Analysis 186
8.7 Maintenance 187
8.8 Summary 188
Mining Actionable Knowledge on Capital Market Data 189
9.1 Case Study 1: Extracting Actionable Trading Strategies 189
9.2 Case Study 2: Mining Actionable Market Microstructure Behavior Patterns 204
Mining Actionable Knowledge on Social Security Data 210
10.1 Case Study: Mining Actionable Combined Associations 210
10.2 Experiments: Mining Actionable Combined Patterns 214
10.3 Summary 222
Open Issues and Prospects 223
11.1 Open Issues 223
11.2 Trends and Prospects 224
Reading Materials 226
12.1 Activities on D3M 226
12.2 References on D3M 227
12.3 References on Agent Mining 228
12.4 References on Post-analysis and Post-mining 228
Glossary 229
Reference 237
Index 248