Bayesian Networks, the result of the convergence of artificial intelligence with statistics, are growing in popularity. Their versatility and modelling power is now employed across a variety of fields for the purposes of analysis, simulation, prediction and diagnosis.This book provides a general introduction to Bayesian networks, defining and illustrating the basic concepts with pedagogical examples and twenty real-life case studies drawn from a range of fields including medicine, computing, natural sciences and engineering.Designed to help analysts, engineers, scientists and professionals taking part in complex decision processes to successfully implement Bayesian networks, this book equips readers with proven methods to generate, calibrate, evaluate and validate Bayesian networks.The book:Provides the tools to overcome common practical challenges such as the treatment of missing input data, interaction with experts and decision makers, determination of the optimal granularity and size of the model. Highlights the strengths of Bayesian networks whilst also presenting a discussion of their limitations.Compares Bayesian networks with other modelling techniques such as neural networks, fuzzy logic and fault trees.Describes, for ease of comparison, the main features of the major Bayesian network software packages: Netica, Hugin, Elvira and Discoverer, from the point of view of the user.Offers a historical perspective on the subject and analyses future directions for research.Written by leading experts with practical experience of applying Bayesian networks in finance, banking, medicine, robotics, civil engineering, geology, geography, genetics, forensic science, ecology, and industry, the book has much to offer both practitioners and researchers involved in statistical analysis or modelling in any of these fields.
Author(s): Olivier Pourret, Patrick Naïm, Bruce Marcot
Edition: 1
Publisher: Wiley
Year: 2008
Language: English
Pages: 448
Tags: Математика;Теория вероятностей и математическая статистика;Математическая статистика;Прикладная математическая статистика;
Bayesian Networks......Page 3
Contents......Page 7
Foreword......Page 11
Preface......Page 13
1.1 Models......Page 19
1.2 Probabilistic vs. deterministic models......Page 23
1.3 Unconditional and conditional independence......Page 27
1.4 Bayesian networks......Page 29
2.1 Bayesian networks in medicine......Page 33
2.2 Context and history......Page 35
2.3 Model construction......Page 37
2.4 Inference......Page 44
2.5 Model validation......Page 46
2.6 Model use......Page 48
2.7 Comparison to other approaches......Page 49
2.8 Conclusions and perspectives......Page 50
3.1 Introduction......Page 51
3.2 Models and methodology......Page 52
3.3 The Busselton network......Page 53
3.4 The PROCAM network......Page 58
3.5 The PROCAM Busselton network......Page 62
3.6 Evaluation......Page 64
3.7 The clinical support tool: TakeHeartII......Page 65
3.8 Conclusion......Page 69
4.1 Introduction......Page 71
4.2 Historical perspectives......Page 72
4.3 Complex traits......Page 74
4.4 Bayesian networks to dissect complex traits......Page 77
4.5 Applications......Page 82
4.6 Future challenges......Page 89
5.1 Introduction......Page 91
5.2 Analysis of the factors affecting crime risk......Page 92
5.3 Expert probabilities elicitation......Page 93
5.4 Data preprocessing......Page 94
5.5 A Bayesian network model......Page 96
5.6 Results......Page 98
5.7 Accuracy assessment......Page 101
5.8 Conclusions......Page 102
6.1 Introduction......Page 105
6.2 An indicator-based analysis......Page 107
6.3 The Bayesian network model......Page 115
6.4 Conclusions......Page 127
7.1 Introduction......Page 131
7.2 Building Bayesian networks for inference......Page 134
7.3 Applications of Bayesian networks in forensic science......Page 138
7.4 Conclusions......Page 144
8.1 Context/history......Page 145
8.2 Model construction......Page 147
8.3 Model calibration, validation and use......Page 154
8.4 Conclusions/perspectives......Page 165
9.1 Mineral potential mapping......Page 167
9.2 Classifiers for mineral potential mapping......Page 169
9.3 Bayesian network mapping of base metal deposit......Page 175
9.4 Discussion......Page 184
9.5 Conclusions......Page 189
10.1 Introduction......Page 191
10.2 Probabilistic relational models......Page 193
10.3 Probabilistic relational student model......Page 194
10.4 Case study......Page 198
10.5 Experimental evaluation......Page 200
10.6 Conclusions and future directions......Page 203
11.1 Introduction......Page 205
11.2 The problem of sensor validation......Page 206
11.3 Sensor validation algorithm......Page 209
11.4 Gas turbines......Page 215
11.5 Models learned and experimentation......Page 216
11.6 Discussion and conclusion......Page 220
12.1 Introduction......Page 221
12.2 Overview......Page 223
12.3 Bayesian networks and information retrieval......Page 224
12.4 Theoretical foundations......Page 225
12.5 Building the information retrieval system......Page 233
12.6 Conclusion......Page 241
13.1 Introduction......Page 243
13.2 Dynamic fault trees......Page 245
13.3 Dynamic Bayesian networks......Page 246
13.4 A case study: The Hypothetical Sprinkler System......Page 248
13.5 Conclusions......Page 255
14 Terrorism risk management......Page 257
14.1 Introduction......Page 258
14.2 The Risk Influence Network......Page 268
14.3 Software implementation......Page 272
14.4 Site Profiler deployment......Page 277
14.5 Conclusion......Page 279
15.1 Introduction......Page 281
15.3 Example of actual credit-ratings systems......Page 282
15.4 Credit-rating data of Japanese companies......Page 284
15.5 Numerical experiments......Page 285
15.6 Performance comparison of classifiers......Page 291
15.7 Conclusion......Page 294
16.1 Introduction......Page 297
16.2 Experimental setup......Page 299
16.3 Feature extraction methods......Page 303
16.4 Classification results......Page 306
16.5 Conclusions......Page 316
17.1 Introduction......Page 319
17.2 Pavement management decisions......Page 320
17.3 Bridge management......Page 325
17.4 Bridge approach embankment – case study......Page 326
17.5 Conclusion......Page 330
18.1 Introduction......Page 331
18.2 A methodology for Root Cause Analysis......Page 332
18.3 Pulp and paper application......Page 339
18.4 The ABB Industrial IT platform......Page 343
18.5 Conclusion......Page 344
19.1 Introduction......Page 347
19.2 Model construction......Page 350
19.3 BayesCredit......Page 353
19.4 Model benchmarking......Page 359
19.5 Benefits from technology and software......Page 360
19.6 Conclusion......Page 361
20.1 Introduction......Page 363
20.2 DeepC......Page 364
20.3 The ADVOCATE II architecture......Page 370
20.4 Model development......Page 372
20.5 Model usage and examples......Page 378
20.6 Benefits from using probabilistic graphical models......Page 379
20.7 Conclusion......Page 380
21.1 Introduction......Page 383
21.2 Human foreknowledge in everyday settings......Page 384
21.3 Machine foreknowledge......Page 387
21.4 Current application and future research needs......Page 391
21.5 Conclusion......Page 393
22.1 An artificial intelligence perspective......Page 395
22.2 A rational approach of knowledge......Page 397
22.3 Future challenges......Page 402
Bibliography......Page 403
Index......Page 445
