Meta-analytics. Consensus approaches and system patterns for data analysis

Cover......Page 1
Meta-Analytics:Consensus Approachesand System Patternsfor Data Analysis......Page 3
Copyright......Page 4
Dedication......Page 5
Acknowledgments......Page 6
 Introduction......Page 8
 Why is this book important?......Page 9
 Organization of the book......Page 10
 Informatics......Page 11
  Value and variance......Page 12
  Sample and population tests......Page 13
  Regression and estimation......Page 17
  k-Means and k-nearest neighbor clustering......Page 23
  Unclustering......Page 26
  Markov models......Page 27
 Machine learning......Page 29
  Entropy......Page 30
  SVM and kernels......Page 31
  Probability......Page 32
  Dimensionality reduction and information gain......Page 34
  Optimization and search......Page 35
  Data mining and knowledge discovery......Page 37
  Recognition......Page 38
  Ensemble learning......Page 40
  Genetic algorithms......Page 43
  Neural networks......Page 49
  Immunological algorithms......Page 56
 A platform for building a classifier from the ground up (binary case)......Page 59
  Training and validation......Page 68
  Testing and deployment......Page 78
  Comparing training and testing data set results......Page 102
 Summary......Page 103
 Further reading......Page 105
 Introduction......Page 106
 Pre-validation......Page 107
 Optimizing settings from training data......Page 115
 Learning how to Learn......Page 121
 Deep learning to deep unlearning......Page 129
 Summary......Page 130
 References......Page 131
 Introduction......Page 132
  Simple (unambiguous) normalization......Page 133
  Bias normalization......Page 134
  Normalization and experimental design tables......Page 139
 Designs for the pruning of aging data......Page 140
  Systems......Page 142
  Hybrid systems......Page 143
  Interfaces......Page 145
  Gain......Page 146
  Domain normalization......Page 149
  Sensitivity analysis......Page 150
 References......Page 151
 Introduction......Page 153
 Cumulative response patterns......Page 154
  Identifying zones of interest......Page 155
  Zones of interest for sequence-dependent predictive selection......Page 158
  Traditional cumulative gain curves, or lift curves......Page 160
  Decision trees......Page 167
  Putative-identity triggered patterns......Page 168
  Expectation-maximization and maximum-minimum patterns......Page 170
 Model agreement patterns......Page 174
  Hybrid regression......Page 175
  Modeling and model fitting......Page 176
 Co-occurrence and similarity patterns......Page 177
 Confusion matrix patterns......Page 179
 Entropy patterns......Page 181
 Independence pattern......Page 184
 Functional NLP patterns (macro-feedback)......Page 188
 Summary......Page 189
 References......Page 190
 Introduction......Page 192
 Sensitivity analysis of the data set itself......Page 195
 Sensitivity analysis of the individual algorithms......Page 199
 Sensitivity analysis of the hybrid algorithmics......Page 201
 Sensitivity analysis of the path to the current state......Page 203
 Summary......Page 205
 References......Page 206
 Introduction......Page 207
 Means of predicting......Page 208
 Means of selecting......Page 210
 Multi-path approach......Page 216
 Applications......Page 218
 Summary......Page 219
 Reference......Page 220
 Introduction......Page 221
 Chemistry analogues for analytics......Page 222
 Organic chemistry analogues for analytics......Page 224
 Immunological and biological analogues for analytics......Page 226
 Anonymization analogues for model design and fitting......Page 228
 LSE, error variance, and entropy: Goodness of fit......Page 229
 Make mine multiple models!......Page 230
 Summary......Page 231
 References......Page 232
 Introduction......Page 233
 Synonym-antonym patterns......Page 234
 Reinforce-void patterns......Page 235
 Broader applicability of these patterns......Page 239
 Further reading......Page 240
 Introduction......Page 241
  Entropy and occurrence vectors......Page 242
  Functional metrics......Page 246
  E-M (expectation-maximization) approaches......Page 248
  System design concerns......Page 249
 Optimizing settings from training data......Page 250
 Hybrid methods......Page 251
 Summary......Page 253
 References......Page 254
 Further reading......Page 255
 Introduction......Page 256
  System considerations-Revisiting the system gains......Page 257
  System gains-Revisiting and expanding the system biases......Page 259
 Module optimization......Page 263
 Clustering and regularization......Page 264
  Sum of squares regularization......Page 268
  Variance regularization......Page 269
  Cluster size regularization......Page 270
  Number of clusters regularization......Page 271
  Discussion of regularization methods......Page 272
 Analytic system optimization......Page 273
 References......Page 274
 Introduction......Page 275
  Sequential removal of features aleatory pattern......Page 277
  Sequential variation of feature output aleatory pattern......Page 280
 Adding random elements for testing......Page 282
 Hyperspectral aleatory approaches......Page 284
 Other aleatory applications in machine and statistical learning......Page 285
 Summary......Page 286
 Further reading......Page 287
 Introduction......Page 288
 Machine translation......Page 289
 Robotics......Page 292
 Biological sciences......Page 296
 Summary......Page 298
 References......Page 299
 Introduction......Page 300
 Healthcare......Page 301
 Economics......Page 303
 Business and finance......Page 306
 Postscript: Psychology......Page 309
 References......Page 311
 Chapter 1......Page 312
 Chapter 2......Page 313
 Chapter 4......Page 314
 Chapter 5......Page 315
 Chapter 7......Page 316
 Chapter 9......Page 317
 Chapter 10......Page 318
 Chapter 12......Page 319
 The future of meta-analytics......Page 320
 C......Page 321
 E......Page 322
 L......Page 323
 P......Page 324
 S......Page 325
 Z......Page 326
Back Cover......Page 327