Case-Based Reasoning Research and Development: 8th International Conference on Case-Based Reasoning, ICCBR 2009 Seattle, WA, USA, July 20-23, 2009 Proceedings

front-matter.pdf......Page 2
Retrieval Knowledge......Page 13
Representation Knowledge......Page 14
Agile CBR......Page 15
Conclusions......Page 16
The Gray Cygnet Problem: Four Research Issues......Page 18
Examples of Black Swans and Gray Cygnets......Page 20
Recognition of Swans and Vigilant Monitoring for Cygnets......Page 21
Hypotheticals and Synthetic Cygnets......Page 22
Responsive Re-representation......Page 23
Introduction......Page 26
Related Work......Page 27
Case Retrieval Net (CRN)......Page 28
From CRN to CR2N......Page 30
Text Reuse with Case Grouping......Page 31
Text Reuse with CR2N......Page 32
Distinguishing CR2N from CG......Page 33
Evaluation Methodology......Page 34
Weather Forecast Revision......Page 35
Health and Safety Incident Reporting......Page 37
Conclusions and Future Work......Page 39
Introduction......Page 41
Transfer Learning and Case-Based Reasoning......Page 42
Case Study: Intent Recognition for Transfer Learning......Page 44
Environment, Tasks, and State Representations......Page 45
Learning Algorithms......Page 47
Empirical Evaluation......Page 48
Results and Analysis......Page 49
A Concurrent Learning Alternative......Page 50
Discussion: Intent Recognition, TL, CBR, and Future Work......Page 52
Summary......Page 53
References......Page 54
Introduction......Page 57
Reasoning with Hypothetical Cases and Adaptation......Page 58
Example of Reasoning with Hypotheticals......Page 59
Case-Based Adaptation in the Process Model......Page 61
Representing Hypothetical Reasoning Diagrammatically......Page 64
Experiment to Assess Reliability of Interpreting Diagrams......Page 66
Experimental Procedure......Page 67
Preliminary Results and Discussion......Page 68
References......Page 70
Introduction......Page 72
Representation Issues......Page 73
The CBR Process in Taaable......Page 75
Why Learning Adaptation Knowledge in Taaable?......Page 76
Adaptation Knowledge Discovery from the Case Base......Page 77
Combining the Two Approaches......Page 78
AK Discovery......Page 79
Opportunistic Adaptation Knowledge Discovery......Page 80
A First Example: Cooking a Chocolate Cake......Page 81
A Second Example: Cooking a Chinese Soup......Page 82
Discussion and Related Work......Page 83
Conclusion and Future Work......Page 85
Introduction......Page 87
Reinforcement Learning and the Q–Learning Algorithm......Page 88
Heuristic Accelerated Reinforcement Learning and the HAQL Algorithm......Page 90
Case Based Reasoning......Page 91
Combining Case Based Reasoning and Reinforcement Learning......Page 93
Experiments in the Robotic Soccer Domain......Page 95
Conclusion......Page 99
Introduction......Page 102
Quality of the CBR System's Knowledge Base......Page 103
Case-Based Reasoner Quality Management......Page 104
A General Framework for CBRQM......Page 105
Basic Conditions and Premises......Page 106
Procedural Elements......Page 107
Behavioural Elements......Page 109
Integrating CBRM and CBRQM......Page 110
Conclusion......Page 113
Introduction of the Running Example......Page 117
Metric Spaces......Page 118
CBR: Definitions and Hypotheses......Page 119
Formalization of the Example......Page 121
IC Merging in Propositional Logic......Page 122
Generalization......Page 123
$\delta$-Combination of Cases......Page 124
Application to the Example......Page 125
Credible Case-Based Inference......Page 126
Computing IC Merging in Numerical Spaces......Page 127
Conclusion, Related Work, and Future Work......Page 128
Introduction......Page 132
Empirical Comparisons......Page 133
An Analysis of Algorithm Biases......Page 134
Maintenance by a Committee of Experts (MACE)......Page 136
Experimental Methodology......Page 138
Harmonic Mean......Page 139
Pareto Front......Page 140
Noise-Filtering......Page 141
The Effect of Boundary Complexity......Page 143
The Special Case of Spam......Page 144
Conclusions and Future Work......Page 145
Introduction......Page 147
Case-Base Editing......Page 148
Competence-Based Case-Base Editing......Page 149
Case Profiles......Page 150
Enhanced Competence Model......Page 151
Categorising Cases......Page 152
Removal of Different Types of Cases......Page 153
What Existing Noise Reduction Algorithms Do......Page 156
Comparison of Editing Algorithms......Page 157
Conclusions and Future Work......Page 159
Introduction......Page 162
Related Work......Page 164
Modelling an Expert's Behaviour......Page 165
Improving Passive Learning with Active Case Generation......Page 166
Determining a Connecting Sequence......Page 168
Experimental Results......Page 169
Experimental Setup......Page 170
Importance of Problem Order......Page 172
Applying Active Learning......Page 173
Conclusions and Future Work......Page 174
Introduction......Page 177
Background......Page 178
Determining Confidence in Candidate Suggestions......Page 179
Determining Relevance of Overlapping Suggestions......Page 180
A General Framework for Producing and Combining Adaptation Suggestions......Page 181
Component Suggestion Methods......Page 182
Methods for Combining Suggestions......Page 183
General Factors and Trade-Offs......Page 185
Results......Page 186
Future Work......Page 189
Conclusions......Page 190
Introduction......Page 192
Analyses of Boundary Conditions......Page 193
Analysis of an Omniscient Adaptation Algorithm......Page 194
A Plan Adaptation Example......Page 195
Domain-Configurable Plan Adaptation......Page 196
Partial-Order Planning......Page 197
Domain-Configurable Partial-Order Plan Adaptation Knowledge......Page 198
Domain-Configurable Partial-Order Plan Adaptation Algorithm......Page 200
Example of Domain-Configurable Plan Adaptation......Page 201
Experimental Setup......Page 202
Results......Page 203
Conclusions......Page 204
References......Page 205
Introduction......Page 207
Boosting CBR Agents......Page 208
Learning Boosting Weights with GA......Page 211
Fitness Function......Page 212
Crossover......Page 213
Reinsertion......Page 214
Breast Cancer Case Base......Page 215
Experimental Set Up......Page 216
Results......Page 217
Conclusions......Page 219
Introduction......Page 222
Related Work......Page 223
Case-Based Planning in WARGUS......Page 224
Meta-level Plan Adaptation......Page 227
Trace Recording......Page 228
Trace Difference Calculation......Page 229
Failure Detection......Page 230
Plan Modification......Page 231
Exemplification......Page 233
Meta-level Plan Adaptation Results......Page 234
Conclusion......Page 235
Introduction......Page 237
Data Structures and Functions for Multi-level Abstractions and Flexible Querying......Page 239
Multi-dimensional Index Structures for Retrieval Optimization......Page 243
Index Generation and Navigation......Page 245
Comparisons with Related Work......Page 248
Conclusions......Page 249
Introduction......Page 252
A Refinement Lattice for Feature Logics......Page 253
Anti-unification-Based Similarity......Page 256
An Illustrative Example......Page 258
Property-Based Similarity Definition......Page 260
Experimental Results......Page 262
Related Work......Page 264
Conclusions......Page 265
Introduction......Page 268
Background......Page 269
The Design Principles of the DDAM Model......Page 270
Dynamicity and Similarity-Based Organization......Page 271
Existing Approaches......Page 272
The Generalized Trie Memory Model......Page 273
Unsupervised Grammar Induction......Page 275
The “Hyperspace Telescope” Analogy......Page 276
A Medical Terminology Experiment......Page 278
Unsupervised Equivalence Set Induction......Page 279
Limitations and Future Work......Page 280
References......Page 281
Introduction......Page 282
Complexity Measures in TCBR......Page 284
Pitfalls in Earlier Approaches......Page 285
Empirical Demonstration......Page 286
Calculating Complexity......Page 287
Correlation Method......Page 288
Visualization Using MST......Page 290
Synthetic Datasets......Page 292
Deerwester Dataset......Page 293
20 Newsgroups Dataset......Page 294
Conclusions......Page 295
Relation to Previous Work......Page 297
Dynamic Reinforcement Learning Problem Statement......Page 298
S-Learning Algorithm......Page 299
Robot Simulation......Page 300
Results......Page 305
Limitations of S-Learning......Page 307
Implications......Page 308
Introduction......Page 310
Background......Page 312
Compatibility Using Reinforcement Learning......Page 313
Similarity Using User Preference Weighting......Page 316
Global User Preference Weighting......Page 317
Setup......Page 318
Reinforcement Learning Recommendation Efficiency......Page 319
Quality Recommendation Efficiency......Page 321
Conclusions......Page 323
Introduction......Page 325
Abstraction in Description Logics......Page 326
The Pizza Restaurant Domain Example......Page 328
Generative Planning Using DLs......Page 329
Planning in the Pizza Restaurant Domain......Page 331
Creating Abstract Cases......Page 332
Case Base Indexing and Retrieval......Page 333
Case Reuse......Page 334
Operator Semantics......Page 335
Related Work and Conclusions......Page 337
Introduction......Page 340
Motivation......Page 341
Related Work......Page 342
The LSVM Noise Reduction Technique......Page 344
Making LSVM Noise Reduction Scalable for Large Datasets......Page 346
Computational Complexity of FaLKNR......Page 349
Experimental Procedure......Page 350
Results and Discussion......Page 351
Conclusions......Page 353
Introduction......Page 355
Nearest Neighbor Retrieval......Page 356
Conceptual Neighborhoods......Page 357
Case Retrieval......Page 358
Retrieval in Concept Space......Page 359
Conceptual Neighborhoods in RTS Games......Page 360
Recall Methods......Page 361
Case Selection......Page 362
Implementation......Page 363
Architecture......Page 364
Results......Page 365
Related Work......Page 367
Conclusion......Page 368
Introduction......Page 370
Decision Model for Decision Analysis......Page 371
Case-Based Learning of Decision Trees......Page 372
Reasoning Degrees of Belief Using the D-S Theory......Page 374
Reaching Final Probabilities via Probabilistic Reasoning......Page 377
Decision Analysis Using Case-Based Decision Model......Page 379
When Utility Values from Cases Are Fuzzy......Page 380
Related Work......Page 382
References......Page 383
Introduction......Page 386
Case-Based Planning......Page 387
Learning for Planning......Page 388
Problem Formulation......Page 389
Encoding Planning Problem......Page 390
Building Constraints......Page 391
Assigning Weights......Page 393
Obtaining a Final Solution Plan......Page 394
Experiment Results......Page 395
Different Number of Cases......Page 396
Different Values of......Page 397
Conclusion......Page 398
Introduction......Page 401
Preliminary Work......Page 402
Generalization......Page 403
Experiments towards an Extensions of the Initial Scenario......Page 404
Results......Page 406
Evaluation of the Experiments and Their Results......Page 409
Related Work......Page 410
Conclusion and Outlook......Page 411
Introduction......Page 415
Episodic Memory Constraints......Page 416
Integration of Episodic Memory in Soar......Page 417
Episodic Memory Integration......Page 418
Evaluation Domain......Page 419
Global Episodic Memory Structures......Page 420
Episode Storage......Page 421
Cue Matching......Page 422
Episode Reconstruction......Page 426
Conclusion......Page 427
Introduction......Page 430
Background Work......Page 431
Case Retrieval from a Clustered Case Memory......Page 432
The jCOLIBRI Framework......Page 434
Connectors......Page 435
In-Memory Organization......Page 436
Execution of Case Retrieval Strategies......Page 437
Textual CBR......Page 439
Organization Based on Self-Organizing Maps......Page 440
Experiments, Results and Discussion......Page 441
Conclusions and Further Work......Page 442
Introduction......Page 446
Maritime Video Surveillance and Related Work......Page 447
The Maritime Activity Analysis Workbench......Page 449
Collective Classification......Page 450
Case-Based Collective Inference......Page 452
Method......Page 456
Discussion......Page 458
Conclusion......Page 459
References......Page 460
Introduction......Page 462
Related Work......Page 464
System Architecture......Page 465
A Hospital Ward Study......Page 466
Test Execution......Page 467
Executing Day 13 Test......Page 468
Executing Day 14 Test......Page 470
Accuracy of the Classifications......Page 471
Analysis and Discussion......Page 472
Conclusion and Further Work......Page 474
Introduction......Page 477
PITS++: Function Value Estimation Prediction......Page 479
Integrated Prediction with CBR......Page 481
Similarity Metric......Page 482
Empirical Evaluation......Page 484
Related Work......Page 487
Conclusions......Page 488
Introduction......Page 491
Current Practices......Page 492
New Technologies......Page 493
A Role for Case-Based Reasoning......Page 494
Cases and Queries......Page 495
Similarity......Page 497
Retrieval and Reuse......Page 498
Experimental Data......Page 499
The Malone Kozak Benchmark System......Page 500
Experimental Methodology......Page 501
Results......Page 502
Conclusions and Future Work......Page 504
Introduction......Page 506
A Motivating Example......Page 507
System Overview......Page 509
Case Bases of Search Experiences......Page 511
On the Creation and Sharing of Search Case Bases......Page 513
Search Collaboration......Page 515
Search Leaders and Followers......Page 516
Promotion Sources......Page 517
Conclusions......Page 518
Introduction......Page 521
The Hole Cleaning Problem......Page 522
KICBR......Page 523
Case Matching......Page 524
Root Causes Assessment......Page 525
Case Matching Results......Page 528
Determining Root Causes of Drilling Problems......Page 531
Conclusion......Page 533
References......Page 534
back-matter.pdf......Page 536