Bisociative Knowledge Discovery: An Introduction to Concept, Algorithms, Tools, and Applications

Author Index
......Page 1
Title
......Page 2
Foreword......Page 5
Part III: Network Analysis
......Page 6
References
......Page 0
Part IV: Exploration
......Page 7
Part V: Applications and Evaluation
......Page 8
Motivation......Page 9
Bisociation......Page 10
Bridging Concepts......Page 11
Bridging Graphs......Page 12
Bridging by Structural Similarity......Page 13
Bisociation Discovery Methods......Page 14
Conclusions......Page 15
References......Page 16
Introduction......Page 19
What Is Creativity?......Page 20
Three Roads to Creativity......Page 22
Koestler\'s Concept of Bisociation......Page 23
Elements of Bisociative Computational Creativity......Page 27
Towards a Formal Definition of Bisociation......Page 32
Related Work......Page 34
Discussion and Conclusion......Page 37
References......Page 39
Introduction......Page 41
Different Categories of Information Network......Page 43
Properties of Relations......Page 44
Ontologies......Page 45
Semantic Networks......Page 46
Topic Maps......Page 47
Weighted Networks......Page 48
BisoNets: Bisociative Information Networks......Page 50
Summary......Page 51
Bridging Concept......Page 52
Bridging Graphs......Page 53
Conclusion......Page 54
References
......Page 55
Network Creation: Overview
......Page 59
References
......Page 61
Introduction......Page 62
Reminder: Bisociation and BisoNets......Page 64
Data Access and Pre-processing......Page 65
Creating Nodes......Page 66
Cosine and Tanimoto Measures......Page 67
The Bison Measure......Page 68
The Probabilistic Measure......Page 69
The Biology and Music Benchmark......Page 70
References......Page 72
Introduction......Page 74
Problem Description......Page 76
Document Acquisition and Preprocessing......Page 78
Document Preprocessing......Page 79
Candidate Concept Detection......Page 81
Distance Measures between Vectors......Page 83
Heuristics Description......Page 85
Tf-idf Based Heuristics......Page 86
Outlier Based Heuristics......Page 87
Baseline Heuristics......Page 88
Evaluation Procedure......Page 89
Migraine-Magnesium Dataset......Page 90
Comparison of the Heuristics......Page 91
Network Creation......Page 94
References
......Page 97
Introduction......Page 99
Related Work......Page 101
The tpf–idf–tpu Model of Important Term Pair Associations......Page 102
Term Pair Uncorrelation (tpu)......Page 103
Experiments......Page 104
Tpf–idf–tpu vs. tf–idf......Page 105
Sentence vs. Document-Level tpf–idf–tpu Methods......Page 106
Comparison of tpf–idf–tpu and tf–idf Using Annotated Test Set......Page 108
Conclusion......Page 109
References
......Page 110
Introduction......Page 112
Frequent Item Set Mining......Page 113
Jaccard Item Sets......Page 116
The Eclat Algorithm......Page 117
The JIM Algorithm (Jaccard Item Set Mining)......Page 118
Other Similarity Measures......Page 119
Experiments......Page 122
References
......Page 127
Introduction......Page 130
Using Graphs: Biomine......Page 131
Using Logic: ProbLog......Page 133
Inference and Reasoning Techniques......Page 136
Deduction: Reasoning about Node Tuples......Page 137
Induction: Finding Patterns......Page 138
Combining Induction and Deduction......Page 140
Modifying the Knowledge Base......Page 141
The Probabilistic Model of Biomine and ProbLog......Page 143
Probabilistic Deduction......Page 146
Patterns and Probabilities......Page 147
Modifying the Probabilistic Knowledge Base......Page 148
Conclusions......Page 149
References
......Page 150
Network Analysis: Overview
......Page 152
References......Page 154
Introduction......Page 155
Motivating Example......Page 156
Requirements......Page 158
Data Representation......Page 160
Basic Data Manipulation......Page 161
Illustrative Examples......Page 164
Databases......Page 168
Conclusions......Page 170
References
......Page 171
Introduction......Page 174
Node Centrality......Page 175
Graph Partitioning......Page 176
Hierarchical Clustering......Page 177
Preference-Dependent Methods......Page 178
Detecting Interesting Nodes or Paths......Page 179
Similarity Subgraph Search......Page 180
Conclusion......Page 181
References
......Page 182
Introduction......Page 187
Definitions......Page 188
Multiplicativity of Ratio of Connectivity Kept......Page 190
A Bound on the Ratio of Connectivity Kept......Page 191
A Further Bound on the Ratio of Connectivity Kept......Page 192
Brute Force Approach......Page 193
Combinational Approach......Page 195
Experiments......Page 196
Results......Page 197
Related Work......Page 203
Conclusion......Page 204
References
......Page 205
Introduction......Page 207
Weighted and Compressed Graphs......Page 210
Simple Weighted Graph Compression......Page 211
Generalized Weighted Graph Compression......Page 212
Optimal Superedge Weights and Mergers......Page 213
Bounds for Distances between Graphs......Page 214
A Bound on Distances between Nodes......Page 215
Related Work......Page 216
Algorithms......Page 217
Experiments......Page 219
Results......Page 220
References
......Page 224
Introduction......Page 226
Related Work......Page 227
Similarities in Probabilistic Graphs......Page 228
Clustering and Representatives in Graphs......Page 229
Test Setting......Page 231
Results......Page 233
Conclusions......Page 235
References
......Page 236
Introduction......Page 238
Bisociative Information Networks......Page 239
Concept Graphs......Page 240
Preliminaries......Page 241
Detection......Page 244
Results......Page 246
Conclusion and Future work......Page 251
References
......Page 252
Introduction......Page 254
Related Work......Page 256
Linear Standard Scenario......Page 257
Node Signatures......Page 258
Node Similarities......Page 259
Activation Similarity......Page 260
Signature Similarity......Page 262
Experiments......Page 263
Schools-Wikipedia......Page 264
Conclusion......Page 268
Motivation......Page 271
Knowledge Modeling......Page 272
Domains......Page 273
Bisociations......Page 275
Finding and Assessing Bisociations......Page 277
Domain Extraction......Page 278
Scoring Bisociation Candidates......Page 280
Complexity and Scalability......Page 281
Preliminary Evaluation......Page 283
Related Work......Page 289
References
......Page 291
Contributions......Page 293
References
......Page 294
Introduction......Page 295
Different Meanings of Exploration......Page 296
Implications for User Interface Design......Page 298
Supporting Bisociative Data Exploration......Page 299
Tools for Data Exploration......Page 300
Evaluation of Knowledge Discovery Tools......Page 302
Open Issues......Page 303
Benchmark Evaluation for Discovery Tools......Page 304
Conclusion and Future Work......Page 305
References
......Page 306
Introduction......Page 309
State of the Art in Graph Interaction and Visualization......Page 310
The Creative Exploration Toolkit......Page 311
The KNIME Information Mining Platform......Page 312
Evaluation......Page 314
Study Design......Page 315
Results of the Study......Page 316
Conclusion and Future Work......Page 319
References
......Page 320
Introduction......Page 321
Related Work in Literature Mining......Page 322
The Upgraded RaJoLink Knowledge Discovery Process......Page 323
Outlier Detection in the RaJoLink Knowledge Discovery Process......Page 325
Application of Outlier Detection in the Autism Literature......Page 327
Conclusions......Page 329
References......Page 330
Introduction......Page 333
Related Work......Page 334
Experimental Datasets......Page 336
Classification Noise Filters for Outlier Detection......Page 338
Experimental Evaluation......Page 339
Conclusions......Page 343
References
......Page 344
Introduction......Page 346
Methodology for Bridging Concept Identification and Ranking......Page 348
Base Heuristics......Page 349
Ensemble Heuristic......Page 351
Experimental Setting......Page 355
Results in the Migraine-Magnesium Dataset......Page 357
Results in Autism-Calcineurin Dataset......Page 359
The CrossBee System
......Page 361
A Typical Use Case......Page 362
Discussion and Further Work......Page 364
References......Page 365
Contributions......Page 367
Application Potential of the BISON Methodology......Page 369
References
......Page 370
Introduction......Page 372
Data Model......Page 373
Source Databases......Page 374
Edge Goodness in Biomine......Page 377
Link Goodness Measures......Page 379
Path and Neighbourhood Level......Page 380
Subgraph Level......Page 381
Graph Level......Page 382
Estimation of Link Significance......Page 383
Conclusion......Page 384
References
......Page 385
Introduction......Page 387
Exploratory Gene Analytics......Page 390
SEGS: Search for Enriched Gene Sets......Page 391
SegMine: Contextualization of genes......Page 392
A Case Study......Page 393
Conclusions......Page 395
References
......Page 396
Introduction......Page 398
Related Work......Page 401
Contrast Mining from Interesting Subgroups......Page 402
Subgroup Discovery (Step 1)......Page 403
Subgroup Discovery (Step 3)......Page 405
An Instance of Our Method: Gene Set Enrichment from Enriched Gene Sets......Page 406
Experiments......Page 408
Conclusion......Page 411
References
......Page 412
Introduction......Page 415
The Probabilistic Logic Environment: ProbLog......Page 417
Metabolic Network Representation......Page 419
Models for Automatic Network Curation......Page 422
Agnostic Noise Model......Page 428
Noise Model for Unreliable Predictions......Page 431
Conclusions......Page 432
References
......Page 433
Introduction......Page 435
Related Work......Page 436
Manually Constructed Petri Net Model of Plant Defence Response......Page 437
Automated Extraction of Plant Defence Response Model from Biological Literature......Page 439
Crossing the Boundaries of Individual Readers......Page 441
Conclusion......Page 443
References
......Page 444
Introduction......Page 446
The Bile Acid and Xenobiotic System......Page 447
Materials and Methods......Page 449
Results and Discussion......Page 456
Conclusions......Page 458
References
......Page 459
Introduction......Page 460
Fuzzy Grammars......Page 463
Fuzzy Formal Concept Analysis......Page 464
Process Data......Page 467
Illustrative Example......Page 470
Business Process Example - Definition of Domains......Page 471
Bisociations......Page 472
Summary......Page 478
References......Page 479
Introduction......Page 480
Interfaces for Creative Music Discovery......Page 482
User-Adaption during the Exploration Process......Page 483
The MusicGalaxy Visualization......Page 484
Orthogonal Similarity Measures......Page 486
Generalization to Domain Graphs......Page 487
Discussion......Page 488
Conclusions......Page 489
References
......Page 490