Hugendubel.info - Die B2B Online-Buchhandlung 

Merkliste
Die Merkliste ist leer.
Bitte warten - die Druckansicht der Seite wird vorbereitet.
Der Druckdialog öffnet sich, sobald die Seite vollständig geladen wurde.
Sollte die Druckvorschau unvollständig sein, bitte schliessen und "Erneut drucken" wählen.

Data Mining and Knowledge Discovery Handbook

E-BookPDF1 - PDF WatermarkE-Book
1285 Seiten
Englisch
SPRINGER USerschienen am10.09.20102nd ed. 2010
This book organizes key concepts, theories, standards, methodologies, trends, challenges and applications of data mining and knowledge discovery in databases. It first surveys, then provides comprehensive yet concise algorithmic descriptions of methods, including classic methods plus the extensions and novel methods developed recently. It also gives in-depth descriptions of data mining applications in various interdisciplinary industries.



Prof. Oded Maimon is the Oracle chaired Professor at Tel-Aviv University, Previously at MIT. Oded is a leader expert in the field of data mining and knowledge discovery. He published many articles on new algorithms and seven significant award winning books in the field since 2000. He has also developed and implemented successful applications in the Industry. He heads an international research group sponsored by European Union awards.

Dr. Lior Rokach is a senior lecturer at the Department of Information System Engineering at Ben-Gurion University. He is a recognized expert in intelligent information systems and has held several leading positions in this field. His main areas of interest are Data Mining, Pattern Recognition, and Recommender Systems. Dr. Rokach is the author of over 70 refereed papers in leading journals, conference proceedings and book chapters. In addition he has authored six books and edited three others books.mehr
Verfügbare Formate
BuchGebunden
EUR374,49
E-BookPDF1 - PDF WatermarkE-Book
EUR229,00
E-BookPDF1 - PDF WatermarkE-Book
EUR394,83

Produkt

KlappentextThis book organizes key concepts, theories, standards, methodologies, trends, challenges and applications of data mining and knowledge discovery in databases. It first surveys, then provides comprehensive yet concise algorithmic descriptions of methods, including classic methods plus the extensions and novel methods developed recently. It also gives in-depth descriptions of data mining applications in various interdisciplinary industries.



Prof. Oded Maimon is the Oracle chaired Professor at Tel-Aviv University, Previously at MIT. Oded is a leader expert in the field of data mining and knowledge discovery. He published many articles on new algorithms and seven significant award winning books in the field since 2000. He has also developed and implemented successful applications in the Industry. He heads an international research group sponsored by European Union awards.

Dr. Lior Rokach is a senior lecturer at the Department of Information System Engineering at Ben-Gurion University. He is a recognized expert in intelligent information systems and has held several leading positions in this field. His main areas of interest are Data Mining, Pattern Recognition, and Recommender Systems. Dr. Rokach is the author of over 70 refereed papers in leading journals, conference proceedings and book chapters. In addition he has authored six books and edited three others books.
Details
Weitere ISBN/GTIN9780387098234
ProduktartE-Book
EinbandartE-Book
FormatPDF
Format Hinweis1 - PDF Watermark
FormatE107
Erscheinungsjahr2010
Erscheinungsdatum10.09.2010
Auflage2nd ed. 2010
Seiten1285 Seiten
SpracheEnglisch
IllustrationenXX, 1285 p. 40 illus.
Artikel-Nr.1540908
Rubriken
Genre9200

Inhalt/Kritik

Inhaltsverzeichnis
1;Preface;6
2;Contents;8
3;List of Contributors;13
4;1 Introduction to Knowledge Discovery and Data Mining;19
4.1;1.1 The KDD Process;20
4.2;1.2 Taxonomy of Data Mining Methods;23
4.3;1.3 Data Mining within the Complete Decision Support System;25
4.4;1.4 KDD and DM Research Opportunities and Challenges;26
4.5;1.5 KDD & DM Trends;27
4.6;1.6 The Organization of the Handbook;28
4.7;1.7 New to This Edition;29
4.7.1;1.7.1 Mining Rich Data Formats;29
4.7.2;1.7.2 New Techniques;30
4.7.3;1.7.3 New Application Domains;30
4.7.4;1.7.4 New Consideration;31
4.7.5;1.7.5 Software;31
4.7.6;1.7.6 Major Updates;31
4.8;References;31
5;Part I Preprocessing Methods;34
5.1;2 Data Cleansing: A Prelude to Knowledge Discovery;35
5.1.1;2.1 INTRODUCTION;35
5.1.2;2.2 DATA CLEANSING BACKGROUND;36
5.1.3;2.3 GENERAL METHODS FOR DATA CLEANSING;39
5.1.4;2.4 APPLYING DATA CLEANSING;40
5.1.4.1;2.4.1 Statistical Outlier Detection;40
5.1.4.2;2.4.2 Clustering;41
5.1.4.3;2.4.3 Pattern-based detection;41
5.1.4.4;2.4.4 Association Rules;42
5.1.5;2.5 CONCLUSIONS;45
5.1.6;References;45
5.2;3 Handling Missing Attribute Values;49
5.2.1;3.1 Introduction;49
5.2.2;3.2 Sequential Methods;51
5.2.2.1;3.2.1 Deleting Cases with Missing Attribute Values;51
5.2.2.2;3.2.2 The Most Common Value of an Attribute;51
5.2.2.3;3.2.3 The Most Common Value of an Attribute Restricted to a Concept;52
5.2.2.4;3.2.4 Assigning All Possible Attribute Values to a Missing Attribute Value;52
5.2.2.5;3.2.5 Assigning All Possible Attribute Values Restricted to a Concept;54
5.2.2.6;3.2.6 Replacing Missing Attribute Values by the Attribute Mean;55
5.2.2.7;3.2.7 Replacing Missing Attribute Values by the Attribute Mean Restricted to a Concept;56
5.2.2.8;3.2.8 Global Closest Fit;56
5.2.2.9;3.2.9 Concept Closest Fit;57
5.2.2.10;3.2.10 Other Methods;58
5.2.3;3.3 Parallel Methods;59
5.2.3.1;3.3.1 Blocks of Attribute-Value Pairs and Characteristic Sets;60
5.2.3.2;3.3.2 Lower and Upper Approximations;62
5.2.3.3;3.3.3 Rule Induction-MLEM2;63
5.2.3.4;3.3.4 Other Approaches to Missing Attribute Values;64
5.2.4;3.4 Conclusions;64
5.2.5;References;64
5.3;4 Geometric Methods for Feature Extraction and Dimensional Reduction - A Guided Tour;68
5.3.1;Introduction;68
5.3.2;4.1 Projective Methods;70
5.3.2.1;4.1.1 Principal Component Analysis (PCA);72
5.3.2.2;4.1.2 Probabilistic PCA (PPCA);76
5.3.2.3;4.1.3 Kernel PCA;77
5.3.2.4;4.1.4 Oriented PCA and Distortion Discriminant Analysis;80
5.3.3;4.2 Manifold Modeling;81
5.3.3.1;4.2.1 The Nystr¨om method;81
5.3.3.2;4.2.2 Multidimensional Scaling;84
5.3.3.3;4.2.3 Isomap;89
5.3.3.4;4.2.4 Locally Linear Embedding;89
5.3.3.5;4.2.5 Graphical Methods;91
5.3.4;4.3 Pulling the Threads Together;93
5.3.5;Acknowledgments;95
5.3.6;References;95
5.4;5 Dimension Reduction and Feature Selection;98
5.4.1;5.1 Introduction;98
5.4.2;5.2 Feature Selection Techniques;101
5.4.2.1;5.2.1 Feature Filters;101
5.4.2.2;5.2.2 Feature Wrappers;106
5.4.3;5.3 Variable Selection;110
5.4.3.1;5.3.1 Mallows Cp (Mallows, 1973);110
5.4.3.2;5.3.2 AIC, BIC and F ratio;111
5.4.3.3;5.3.3 Principal Component Analysis (PCA);111
5.4.3.4;5.3.4 Factor Analysis (FA);112
5.4.3.5;5.3.5 Projection Pursuit;112
5.4.3.6;5.3.6 Advanced Methods for Variable Selection;112
5.4.4;References;112
5.5;6 Discretization Methods;116
5.5.1;Introduction;116
5.5.2;6.1 Terminology;117
5.5.2.1;6.1.1 Qualitative vs. quantitative;117
5.5.2.2;6.1.2 Levels of measurement scales;117
5.5.2.3;6.1.3 Summary;118
5.5.3;6.2 Taxonomy;119
5.5.4;6.3 Typical methods;120
5.5.4.1;6.3.1 Background and terminology;121
5.5.4.2;6.3.2 Equal-width, equal-frequency and fixed-frequency discretization;121
5.5.4.3;6.3.3 Multi-interval-entropy-minimization discretization ((MIEMD);122
5.5.4.4;6.3.4 ChiMerge, StatDisc and InfoMerge discretization;122
5.5.4.5;6.3.5 Cluster-based discretization;123
5.5.4.6;6.3.6 ID3 discretization;123
5.5.4.7;6.3.7 Non-disjoint discretization;123
5.5.4.8;6.3.8 Lazy discretization;124
5.5.4.9;6.3.9 Dynamic-qualitative discretization;125
5.5.4.10;6.3.10 Ordinal discretization;125
5.5.4.11;6.3.11 Fuzzy discretization;125
5.5.4.12;6.3.12 Iterative-improvement discretization;126
5.5.4.13;6.3.13 Summary;126
5.5.5;6.4 Discretization and the learning context;126
5.5.5.1;6.4.1 Discretization for decision tree learning;127
5.5.5.2;6.4.2 Discretization for naive-Bayes learning;127
5.5.6;6.5 Summary;128
5.5.7;References;129
5.6;7 Outlier Detection;132
5.6.1;7.1 Introduction: Motivation, Definitions and Applications;132
5.6.2;7.2 Taxonomy of Outlier Detection Methods;133
5.6.3;7.3 Univariate Statistical Methods;134
5.6.3.1;7.3.1 Single-step vs. Sequential Procedures;134
5.6.3.2;7.3.2 Inward and Outward Procedures;135
5.6.3.3;7.3.3 Univariate Robust Measures;135
5.6.3.4;7.3.4 Statistical Process Control (SPC);136
5.6.4;7.4 Multivariate Outlier Detection;137
5.6.4.1;7.4.1 Statistical Methods for Multivariate Outlier Detection;138
5.6.4.2;7.4.2 Multivariate Robust Measures;139
5.6.4.3;7.4.3 Data-Mining Methods for Outlier Detection;139
5.6.4.4;7.4.4 Preprocessing Procedures;141
5.6.5;7.5 Comparison of Outlier Detection Methods;141
5.6.6;References;142
6;Part II Supervised Methods;146
6.1;8 Supervised Learning;147
6.1.1;8.1 Introduction;147
6.1.2;8.2 Training Set;148
6.1.3;8.3 Definition of the Classification Problem;148
6.1.4;8.4 Induction Algorithms;149
6.1.5;8.5 Performance Evaluation;150
6.1.5.1;8.5.1 Generalization Error;150
6.1.5.2;8.5.2 Theoretical Estimation of Generalization Error;151
6.1.5.3;8.5.3 Empirical Estimation of Generalization Error;153
6.1.5.4;8.5.4 Computational Complexity;154
6.1.5.5;8.5.5 Comprehensibility;154
6.1.6;8.6 Scalability to Large Datasets;155
6.1.7;8.7 The Curse of Dimensionality ;156
6.1.8;8.8 Classification Problem Extensions;158
6.1.9;References;159
6.2;9 Classification Trees;162
6.2.1;9.1 Decision Trees;162
6.2.2;9.2 Algorithmic Framework for Decision Trees;164
6.2.3;9.3 Univariate Splitting Criteria;164
6.2.3.1;9.3.1 Overview;164
6.2.3.2;9.3.2 Impurity-based Criteria;166
6.2.3.3;9.3.3 Information Gain;166
6.2.3.4;9.3.4 Gini Index;166
6.2.3.5;9.3.5 Likelihood-Ratio Chi-Squared Statistics;167
6.2.3.6;9.3.6 DKM Criterion;167
6.2.3.7;9.3.7 Normalized Impurity Based Criteria;167
6.2.3.8;9.3.8 Gain Ratio;168
6.2.3.9;9.3.9 Distance Measure;168
6.2.3.10;9.3.10 Binary Criteria;168
6.2.3.11;9.3.11 Twoing Criterion;168
6.2.3.12;9.3.12 Orthogonal (ORT) Criterion;169
6.2.3.13;9.3.13 Kolmogorov-Smirnov Criterion;169
6.2.3.14;9.3.14 AUC-Splitting Criteria;169
6.2.3.15;9.3.15 Other Univariate Splitting Criteria;169
6.2.3.16;9.3.16 Comparison of Univariate Splitting Criteria;170
6.2.4;9.4 Multivariate Splitting Criteria;170
6.2.5;9.5 Stopping Criteria;170
6.2.6;9.6 Pruning Methods;171
6.2.6.1;9.6.1 Overview;171
6.2.6.2;9.6.2 Cost-Complexity Pruning;171
6.2.6.3;9.6.3 Reduced Error Pruning;172
6.2.6.4;9.6.4 Minimum Error Pruning (MEP);172
6.2.6.5;9.6.5 Pessimistic Pruning;172
6.2.6.6;9.6.6 Error-based Pruning (EBP);173
6.2.6.7;9.6.7 Optimal Pruning;173
6.2.6.8;9.6.8 Minimum Description Length (MDL) Pruning;174
6.2.6.9;9.6.9 Other Pruning Methods;174
6.2.6.10;9.6.10 Comparison of Pruning Methods;174
6.2.7;9.7 Other Issues;175
6.2.7.1;9.7.1 Weighting Instances;175
6.2.7.2;9.7.2 Misclassification costs;175
6.2.7.3;9.7.3 Handling Missing Values;175
6.2.8;9.8 Decision Trees Inducers;176
6.2.8.1;9.8.1 ID3;176
6.2.8.2;9.8.2 C4.5;176
6.2.8.3;9.8.3 CART;177
6.2.8.4;9.8.4 CHAID;177
6.2.8.5;9.8.5 QUEST;178
6.2.8.6;9.8.6 Reference to Other Algorithms;178
6.2.9;9.9 Advantages and Disadvantages of Decision Trees;178
6.2.10;9.10 Decision Tree Extensions;180
6.2.10.1;9.10.1 Oblivious Decision Trees;180
6.2.10.2;9.10.2 Fuzzy Decision Trees;181
6.2.10.3;9.10.3 Decision Trees Inducers for Large Datasets;182
6.2.10.4;9.10.4 Incremental Induction;182
6.2.11;References;183
6.3;10 Bayesian Networks;188
6.3.1;10.1 Introduction;188
6.3.2;10.2 Representation;189
6.3.3;10.3 Reasoning;192
6.3.4;10.4 Learning;194
6.3.4.1;10.4.1 Scoring Metrics;194
6.3.4.2;10.4.2 Model Search;201
6.3.4.3;10.4.3 Validation;202
6.3.5;10.5 Bayesian Networks in Data Mining;204
6.3.5.1;10.5.1 Bayesian Networks and Classification;204
6.3.5.2;10.5.2 Generalized Gamma Networks;206
6.3.5.3;10.5.3 Bayesian Networks and Dynamic Data;208
6.3.6;10.6 Data Mining Applications;211
6.3.6.1;10.6.1 Survey Data;211
6.3.6.2;10.6.2 Customer Profiling;214
6.3.7;10.7 Conclusions and Future Research Directions;216
6.3.8;Acknowledgments;218
6.3.9;References;218
6.4;11 Data Mining within a Regression Framework;222
6.4.1;11.1 Introduction;222
6.4.2;11.2 Some Definitions;223
6.4.3;11.3 Regression Splines;224
6.4.4;11.4 Smoothing Splines;227
6.4.5;11.5 LocallyWeighted Regression as a Smoother;229
6.4.6;11.6 Smoothers for Multiple Predictors;230
6.4.6.1;11.6.1 The Generalized Additive Model;231
6.4.7;11.7 Recursive Partitioning;233
6.4.7.1;11.7.1 Classification and Regression Trees and Extensions;233
6.4.7.2;11.7.2 Overfitting and Ensemble Methods;239
6.4.8;11.8 Conclusions;242
6.4.9;Acknowledgments;242
6.4.10;References;242
6.5;12 Support Vector Machines;244
6.5.1;12.1 Introduction;244
6.5.2;12.2 Hyperplane Classifiers;245
6.5.2.1;12.2.1 The Linear Classifier;246
6.5.2.2;12.2.2 The Kernel Trick;248
6.5.2.3;12.2.3 The Optimal Margin Support Vector Machine;249
6.5.3;12.3 Non-Separable SVM Models;250
6.5.3.1;12.3.1 Soft Margin Support Vector Classifiers;250
6.5.3.2;12.3.2 Support Vector Regression;252
6.5.3.3;12.3.3 SVM-like Models;254
6.5.4;12.4 Implementation Issues with SVM;254
6.5.4.1;12.4.1 Optimization Techniques;255
6.5.4.2;12.4.2 Model Selection;256
6.5.4.3;12.4.3 Multi-Class SVM;256
6.5.5;12.5 Extensions and Application;257
6.5.6;12.6 Conclusion;258
6.5.7;References;258
6.6;13 Rule Induction;261
6.6.1;13.1 Introduction;261
6.6.2;13.2 Types of Rules;263
6.6.3;13.3 Rule Induction Algorithms;265
6.6.3.1;13.3.1 LEM1 Algorithm;265
6.6.3.2;13.3.2 LEM2;269
6.6.3.3;13.3.3 AQ;272
6.6.4;13.4 Classification Systems;274
6.6.5;13.5 Validation;275
6.6.6;13.6 Advanced Methodology;276
6.6.7;References;276
7;Part III Unsupervised Methods;278
7.1;14 A survey of Clustering Algorithms;279
7.1.1;14.1 Introduction;279
7.1.2;14.2 Distance Measures;280
7.1.2.1;14.2.1 Minkowski: Distance Measures for Numeric Attributes;280
7.1.2.2;14.2.2 Distance Measures for Binary Attributes;281
7.1.2.3;14.2.3 Distance Measures for Nominal Attributes;281
7.1.2.4;14.2.4 Distance Metrics for Ordinal Attributes;281
7.1.2.5;14.2.5 Distance Metrics for Mixed-Type Attributes;282
7.1.3;14.3 Similarity Functions;282
7.1.3.1;14.3.1 Cosine Measure;282
7.1.3.2;14.3.2 Pearson Correlation Measure;283
7.1.3.3;14.3.3 Extended Jaccard Measure;283
7.1.3.4;14.3.4 Dice Coefficient Measure;283
7.1.4;14.4 Evaluation Criteria Measures;283
7.1.4.1;14.4.1 Internal Quality Criteria;283
7.1.4.2;14.4.2 External Quality Criteria;287
7.1.5;14.5 Clustering Methods;288
7.1.5.1;14.5.1 Hierarchical Methods;288
7.1.5.2;14.5.2 Partitioning Methods;290
7.1.5.3;14.5.3 Density-based Methods;292
7.1.5.4;14.5.4 Model-based Clustering Methods;293
7.1.5.5;14.5.5 Grid-based Methods;294
7.1.5.6;14.5.6 Soft-computing Methods;294
7.1.5.7;14.5.7 Which Technique To Use?;298
7.1.6;14.6 Clustering Large Data Sets;299
7.1.6.1;14.6.1 Decomposition Approach;300
7.1.6.2;14.6.2 Incremental Clustering;300
7.1.6.3;14.6.3 Parallel Implementation;302
7.1.7;14.7 Determining the Number of Clusters;302
7.1.7.1;14.7.1 Methods Based on Intra-Cluster Scatter;302
7.1.7.2;14.7.2 Methods Based on both the Inter- and Intra-Cluster Scatter;303
7.1.7.3;14.7.3 Criteria Based on Probabilistic;305
7.1.8;References;305
7.2;15 Association Rules;309
7.2.1;15.1 Introduction;309
7.2.2;15.1.1 Formal Problem Definition;310
7.2.3;15.2 Association Rule Mining;311
7.2.3.1;15.2.1 Association Mining Phase;312
7.2.3.2;15.2.2 Rule Generation Phase;315
7.2.4;15.3 Application to Other Types of Data;317
7.2.5;15.4 Extensions of the Basic Framework;318
7.2.5.1;15.4.1 Some other Rule Evaluation Measures;319
7.2.5.2;15.4.2 Interactive or Knowledge-Based Filtering;320
7.2.5.3;15.4.3 Compressed Representations;321
7.2.5.4;15.4.4 Additional Constraints for Dense Databases;322
7.2.5.5;15.4.5 Rules without Minimum Support;324
7.2.6;15.5 Conclusions;326
7.2.7;References;327
7.3;16 Frequent Set Mining;330
7.3.1;Introduction;330
7.3.2;16.1 Problem Description;330
7.3.3;16.2 Apriori;333
7.3.4;16.3 Eclat;336
7.3.5;16.4 Optimizations;337
7.3.5.1;16.4.1 Item reordering;338
7.3.5.2;16.4.2 Partition;338
7.3.5.3;16.4.3 Sampling;339
7.3.5.4;16.4.4 FP-tree;339
7.3.6;16.5 Concise representations;340
7.3.6.1;16.5.1 Maximal Frequent Sets;340
7.3.6.2;16.5.2 Closed Frequent Sets;341
7.3.6.3;16.5.3 Non Derivable Frequent Sets;342
7.3.7;16.6 Theoretical Aspects;342
7.3.8;16.7 Further Reading;343
7.3.9;References;344
7.4;17 Constraint-based Data Mining;348
7.4.1;17.1 Motivations;348
7.4.2;17.2 Background and Notations;350
7.4.3;17.3 Solving Anti-Monotonic Constraints;353
7.4.4;17.4 Introducing non Anti-Monotonic Constraints;354
7.4.4.1;17.4.1 The Seminal Work;355
7.4.4.2;17.4.2 Generic Algorithms;357
7.4.4.3;17.4.3 Ad-hoc Strategies;359
7.4.4.4;17.4.4 Other Directions of Research;359
7.4.5;17.5 Conclusion;360
7.4.6;References;361
7.5;18 Link Analysis;364
7.5.1;18.1 Introduction;364
7.5.2;18.2 Social Network Analysis;366
7.5.3;18.3 Search Engines;368
7.5.4;18.4 Viral Marketing;370
7.5.5;18.5 Law Enforcement & Fraud Detection;372
7.5.6;18.6 Combining with Traditional Methods;374
7.5.7;18.7 Summary;375
7.5.8;References;375
8;Part IV Soft Computing Methods;378
8.1;19 A Review of Evolutionary Algorithms for Data Mining;379
8.1.1;19.1 Introduction;379
8.1.2;19.2 An Overview of Evolutionary Algorithms;380
8.1.3;19.3 Evolutionary Algorithms for Discovering Classification Rules;382
8.1.3.1;19.3.1 Individual Representation for Classification-Rule Discovery;382
8.1.3.2;19.3.2 Searching for a Diverse Set of Rules;385
8.1.3.3;19.3.3 Fitness Evaluation;386
8.1.4;19.4 Evolutionary Algorithms for Clustering;389
8.1.4.1;19.4.1 Individual Representation for Clustering;389
8.1.4.2;19.4.2 Fitness Evaluation for Clustering;391
8.1.5;19.5 Evolutionary Algorithms for Data Preprocessing;392
8.1.5.1;19.5.1 Genetic Algorithms for Attribute Selection;392
8.1.5.2;19.5.2 Genetic Programming for Attribute Construction;394
8.1.6;19.6 Multi-Objective Optimization with Evolutionary Algorithms;397
8.1.7;19.7 Conclusions;401
8.1.8;References;403
8.2;20 A Review of Reinforcement Learning Methods;409
8.2.1;20.1 Introduction;409
8.2.2;20.2 The Reinforcement-Learning Model;410
8.2.3;20.3 Reinforcement-Learning Algorithms;412
8.2.3.1;20.3.1 Dynamic-Programming;412
8.2.3.2;20.3.2 Generalization of Dynamic-Programming to Reinforcement-Learning;413
8.2.4;20.4 Extensions to Basic Model and Algorithms;416
8.2.4.1;20.4.1 Multi-Agent RL;416
8.2.4.2;20.4.2 Tackling Large Sets of States and Actions;417
8.2.5;20.5 Applications of Reinforcement-Learning;417
8.2.6;20.6 Reinforcement-Learning and Data-Mining;418
8.2.7;20.7 An Instructive Example;419
8.2.8;References;422
8.3;21 Neural Networks For Data Mining;426
8.3.1;21.1 Introduction;426
8.3.2;21.2 A Brief History;427
8.3.3;21.3 Neural Network Models;429
8.3.3.1;21.3.1 Feedforward Neural Networks;429
8.3.3.2;21.3.2 Hopfield Neural Networks;438
8.3.3.3;21.3.3 Kohonen´s Self-organizing Maps;440
8.3.4;21.4 Data Mining Applications;443
8.3.5;21.5 Conclusions;445
8.3.6;References;446
8.4;22 Granular Computing and Rough Sets - An Incremental Development;452
8.4.1;22.1 Introduction;452
8.4.2;22.2 Naive Model for Problem Solving;453
8.4.2.1;22.2.1 Information Granulations/Partitions;453
8.4.2.2;22.2.2 Knowledge Level Processing and Computing with Words;454
8.4.2.3;22.2.3 Information Integration and Approximation Theory;454
8.4.3;22.3 A Geometric Models of Information Granulations;455
8.4.4;22.4 Information Granulations/Partitions;456
8.4.4.1;22.4.1 Equivalence Relations(Partitions);456
8.4.4.2;22.4.2 Binary Relation (Granulation) - Topological Partitions;457
8.4.4.3;22.4.3 Fuzzy Binary Granulations (Fuzzy Binary Relations);457
8.4.5;22.5 Non-partition Application - Chinese Wall Security Policy Model;457
8.4.5.1;22.5.1 Simple Chinese Wall Security Policy;458
8.4.6;22.6 Knowledge Representations;459
8.4.6.1;22.6.1 Relational Tables and Partitions;459
8.4.6.2;22.6.2 Table Representations of Binary Relations;460
8.4.6.3;22.6.3 New representations of topological relations;463
8.4.7;22.7 Topological Concept Hierarchy Lattices/Trees;464
8.4.7.1;22.7.1 Granular Lattice;464
8.4.7.2;22.7.2 Granulated/Quotient Sets;466
8.4.7.3;22.7.3 Tree of centers;466
8.4.7.4;22.7.4 Topological tree;467
8.4.7.5;22.7.5 Table Representation of Fuzzy Binary Relations;468
8.4.8;22.8 Knowledge Processing;469
8.4.8.1;22.8.1 The Notion of Knowledge;469
8.4.8.2;22.8.2 Strong,Weak and Knowledge Dependence;470
8.4.8.3;22.8.3 Knowledge Views of Binary Granulations;470
8.4.9;22.9 Information Integration;471
8.4.9.1;22.9.1 Extensions;471
8.4.9.2;22.9.2 Approximations in Rough Set Theory (RST);471
8.4.9.3;22.9.3 Binary Neighborhood System Spaces;472
8.4.10;22.10 Conclusions;473
8.4.11;References;473
8.5;23 Pattern Clustering Using a Swarm Intelligence Approach;476
8.5.1;23.1 Introduction;476
8.5.2;23.2 An Introduction to Swarm Intelligence;478
8.5.3;23.2.1 The Ant Colony Systems;480
8.5.4;23.3 Data Clustering - An Overview;485
8.5.5;23.3.1 Problem Definition;485
8.5.6;23.3.2 The Classical Clustering Algorithms;486
8.5.7;23.3.3 Relevance of SI Algorithms in Clustering;488
8.5.8;23.4 Clustering with the SI Algorithms;488
8.5.8.1;23.4.1 The Ant Colony Based Clustering Algorithms;488
8.5.8.2;23.4.2 The PSO-based Clustering Algorithms;490
8.5.9;23.5 Automatic Kernel-based Clustering with PSO;492
8.5.9.1;23.5.1 The Kernel Based Similarity Measure;493
8.5.9.2;23.5.2 Reformulation of CS Measure;494
8.5.9.3;23.5.3 The Multi-Elitist PSO (MEPSO) Algorithm;495
8.5.9.4;23.5.4 Particle Representation;496
8.5.9.5;23.5.5 The Fitness Function;497
8.5.9.6;23.5.6 Avoiding Erroneous particles with Empty Clusters or Unreasonable Fitness Evaluation;497
8.5.9.7;23.5.7 Putting It All Together;498
8.5.9.8;23.5.8 Experimental Results;498
8.5.10;23.6 Conclusion and Future Directions;503
8.5.11;References;507
8.6;24 Using Fuzzy Logic in Data Mining;512
8.6.1;24.1 Introduction;512
8.6.2;24.2 Basic Concepts of Fuzzy Set Theory;512
8.6.2.1;24.2.1 Membership function;513
8.6.2.2;24.2.2 Fuzzy Set Operations;515
8.6.3;24.3 Fuzzy Supervised Learning;516
8.6.3.1;24.3.1 Growing Fuzzy Decision Tree;516
8.6.3.2;24.3.2 Soft Regression;521
8.6.3.3;24.3.3 Neuro-fuzzy;521
8.6.4;24.4 Fuzzy Clustering;521
8.6.5;24.5 Fuzzy Association Rules;523
8.6.6;24.6 Conclusion;525
8.6.7;References;525
9;Part V Supporting Methods;528
9.1;25 Statistical Methods for Data Mining;529
9.1.1;25.1 Introduction;529
9.1.2;25.2 Statistical Issues in DM;530
9.1.2.1;25.2.1 Size of the Data and Statistical Theory;530
9.1.2.2;25.2.2 The Curse of Dimensionality and Approaches to Address It;531
9.1.2.3;25.2.3 Assessing Uncertainty;532
9.1.2.4;25.2.4 Automated Analysis;532
9.1.2.5;25.2.5 Algorithms for Data Analysis in Statistics;533
9.1.2.6;25.2.6 Visualization;533
9.1.2.7;25.2.7 Scalability;534
9.1.2.8;25.2.8 Sampling;534
9.1.3;25.3 Modeling Relationships using Regression Models;535
9.1.3.1;25.3.1 Linear Regression Analysis;535
9.1.3.2;25.3.2 Generalized Linear Models;536
9.1.3.3;25.3.3 Logistic Regression;537
9.1.3.4;25.3.4 Survival Analysis;538
9.1.4;25.4 False Discovery Rate (FDR) Control in Hypotheses Testing;539
9.1.5;25.5 Model (Variables or Features) Selection using FDR Penalization in GLM;542
9.1.6;25.6 Concluding Remarks;543
9.1.7;References;544
9.2;26 Logics for Data Mining;547
9.2.1;Introduction;547
9.2.2;26.1 Generalized quantifiers;548
9.2.3;26.2 Some important classes of quantifiers;550
9.2.3.1;26.2.1 One-dimensional;550
9.2.3.2;26.2.2 Two-dimensional;551
9.2.4;26.3 Some comments and conclusion;554
9.2.5;Acknowledgments;555
9.2.6;References;555
9.3;27 Wavelet Methods in Data Mining;558
9.3.1;27.1 Introduction;558
9.3.2;27.2 A Framework for Data Mining Process;559
9.3.3;27.3 Wavelet Background;559
9.3.3.1;27.3.1 Basics of Wavelet in L2(R);559
9.3.3.2;27.3.2 Dilation Equation;560
9.3.3.3;27.3.3 Multiresolution Analysis (MRA) and Fast DWT Algorithm;561
9.3.3.4;27.3.4 Illustrations of Harr Wavelet Transform;562
9.3.3.5;27.3.5 Properties of Wavelets;563
9.3.4;27.4 Data Management;564
9.3.5;27.5 Preprocessing;564
9.3.5.1;27.5.1 Denoising;565
9.3.5.2;27.5.2 Data Transformation;566
9.3.5.3;27.5.3 Dimensionality Reduction;566
9.3.6;27.6 Core Mining Process;567
9.3.6.1;27.6.1 Clustering;567
9.3.6.2;27.6.2 Classification;568
9.3.6.3;27.6.3 Regression;568
9.3.6.4;27.6.4 Distributed Data Mining;569
9.3.6.5;27.6.5 Similarity Search/Indexing;570
9.3.6.6;27.6.6 Approximate Query Processing;571
9.3.6.7;27.6.7 Traffic Modeling;572
9.3.7;27.7 Conclusion;573
9.3.8;References;574
9.4;28 Fractal Mining - Self Similarity-based Clustering and its Applications;577
9.4.1;28.1 Introduction;577
9.4.2;28.2 Fractal Dimension;578
9.4.3;28.3 Clustering Using the Fractal Dimension;582
9.4.3.1;28.3.1 FC Initialization Step;582
9.4.3.2;28.3.2 Incremental Step;582
9.4.3.3;28.3.3 Reshaping Clusters in Mid-Flight;584
9.4.3.4;28.3.4 Complexity of the Algorithm;585
9.4.3.5;28.3.5 Confidence Bounds;585
9.4.3.6;28.3.6 Memory Management;586
9.4.3.7;28.3.7 Experimental Results;587
9.4.4;28.4 Projected Fractal Clustering;589
9.4.5;28.5 Tracking Clusters;590
9.4.5.1;28.5.1 Experiment on a Real Dataset;590
9.4.6;28.6 Conclusions;591
9.4.7;References;592
9.5;29 Visual Analysis of Sequences Using Fractal Geometry;594
9.5.1;29.1 Introduction;594
9.5.2;29.2 Iterated Function System (IFS);595
9.5.3;29.3 Algorithmic Framework;597
9.5.3.1;29.3.1 Overview;597
9.5.3.2;29.3.2 Sequence Representation;598
9.5.3.3;29.3.3 Sequence Transformation;598
9.5.3.4;29.3.4 Sequence Pattern Detection;599
9.5.3.5;29.3.5 Sequence Pattern Detection Algorithm Description:;599
9.5.3.6;29.3.6 Classifiers Selection;601
9.5.4;29.4 Fault Sequence Detection Application;602
9.5.5;29.5 Conclusions and Future Research;602
9.5.6;References;603
9.6;30 Interestingness Measures - On Determining What Is Interesting;605
9.6.1;Introduction;605
9.6.2;30.1 Definitions and Notations;606
9.6.3;30.2 Subjective Interestingness;606
9.6.3.1;30.2.1 The Expert-Driven Grammatical Approach;607
9.6.3.2;30.2.2 The Rule-By-Rule Classification Approach;607
9.6.3.3;30.2.3 Interestingness Via What Is Not Interesting Approach;607
9.6.4;30.3 Objective Interestingness;608
9.6.4.1;30.3.1 Ranking Patterns;608
9.6.4.2;30.3.2 Pruning and Application of Constraints;608
9.6.4.3;30.3.3 Summarization of Patterns;609
9.6.5;30.4 Impartial Interestingness;610
9.6.6;30.5 Concluding Remarks;611
9.6.7;References;611
9.7;31 Quality Assessment Approaches in Data Mining;615
9.7.1;Introduction;615
9.7.2;31.1 Data Pre-processing and Quality Assessment;617
9.7.3;31.2 Evaluation of Classification Methods;617
9.7.3.1;31.2.1 Classification Model Accuracy;617
9.7.3.2;31.2.2 Evaluating the Accuracy of Classification Algorithms;619
9.7.3.3;31.2.3 Interestingness Measures of Classification Rules;622
9.7.4;31.3 Association Rules;622
9.7.4.1;31.3.1 Association Rules Interestingness Measures;623
9.7.4.2;31.3.2 Other approaches for evaluating association rules;625
9.7.5;31.4 Cluster Validity;626
9.7.5.1;31.4.1 Fundamental Concepts of Cluster Validity;626
9.7.5.2;31.4.2 External Criteria;628
9.7.5.3;31.4.3 Internal Criteria;630
9.7.5.4;31.4.4 Relative Criteria;631
9.7.5.5;31.4.5 Fuzzy Clustering;637
9.7.5.6;31.4.6 Other Approaches for Cluster Validity;639
9.7.6;References;640
9.8;32 Data Mining Model Comparison;642
9.8.1;32.1 Data Mining and Statistics;642
9.8.2;32.2 Data Mining Model Comparison;643
9.8.3;32.3 Application to Credit Risk Management;647
9.8.4;32.4 Conclusions;654
9.8.5;References;655
9.9;33 Data Mining Query Languages;656
9.9.1;33.1 The Need for Data Mining Query Languages;656
9.9.2;33.2 Supporting Association Rule Mining Processes;657
9.9.3;33.3 A Few Proposals for Association Rule Mining;659
9.9.3.1;33.3.1 MSQL;659
9.9.3.2;33.3.2 MINE RULE;659
9.9.3.3;33.3.3 DMQL;660
9.9.3.4;33.3.4 OLE DB for DM;661
9.9.3.5;33.3.5 A Critical Evaluation;662
9.9.4;33.4 Conclusion;663
9.9.5;References;664
10;Part VI Advanced Methods;666
10.1;34 Mining Multi-label Data;667
10.1.1;34.1 Introduction;667
10.1.2;34.2 Learning;667
10.1.2.1;34.2.1 Problem Transformation;669
10.1.2.2;34.2.2 Algorithm Adaptation;672
10.1.3;34.3 Dimensionality Reduction;673
10.1.3.1;34.3.1 Feature Selection;674
10.1.3.2;34.3.2 Feature Extraction;674
10.1.4;34.4 Exploiting Label Structure;674
10.1.5;34.5 Scaling Up;675
10.1.6;34.6 Statistics and Datasets;676
10.1.7;34.7 Evaluation Measures;677
10.1.7.1;34.7.1 Bipartitions;677
10.1.7.2;34.7.2 Ranking;679
10.1.7.3;34.7.3 Hierarchical;680
10.1.8;34.8 Related Tasks;680
10.1.9;34.9 Multi-Label Data Mining Software;681
10.1.10;References;681
10.2;35 Privacy in Data Mining;686
10.2.1;35.1 Introduction;686
10.2.2;35.2 On the Classification of Protection Procedures;687
10.2.2.1;35.2.1 Computation-Driven Protection Procedures: the Cryptographic Approach;689
10.2.2.2;35.2.2 Data-driven Protection Procedures;690
10.2.3;35.3 Disclosure Risk Measures;690
10.2.3.1;35.3.1 An Scenario for Identity Disclosure;691
10.2.3.2;35.3.2 Measures for Identity Disclosure;692
10.2.4;35.4 Data Protection Procedures;696
10.2.4.1;35.4.1 Perturbative Methods;697
10.2.4.2;35.4.2 Non-perturbative Methods;702
10.2.4.3;35.4.3 Synthetic Data Generators;703
10.2.4.4;35.4.4 k-Anonymity;704
10.2.5;35.5 Information Loss Measures;705
10.2.5.1;35.5.1 Generic Information Loss Measures;705
10.2.5.2;35.5.2 Specific Information Loss Measures;707
10.2.6;35.6 Trade-off and Visualization;707
10.2.6.1;35.6.1 The Score;707
10.2.6.2;35.6.2 R-U Maps;709
10.2.7;35.7 Conclusions;709
10.2.8;Acknowledgements;709
10.2.9;References;709
10.3;36 Meta-Learning - Concepts and Techniques;716
10.3.1;36.1 Introduction;716
10.3.2;36.2 A Meta-Learning Architecture;717
10.3.2.1;36.2.1 Knowledge-Acquisition Mode;718
10.3.2.2;36.2.2 Advisory Mode;718
10.3.3;36.3 Techniques in Meta-Learning;720
10.3.3.1;36.3.1 Dataset Characterization;720
10.3.3.2;36.3.2 Mapping Datasets to Predictive Models;722
10.3.3.3;36.3.3 Learning from Base-Learners;723
10.3.3.4;36.3.4 Inductive Transfer and Learning to Learn;724
10.3.3.5;36.3.5 Dynamic-Bias Selection;725
10.3.4;36.4 Tools and Applications;725
10.3.4.1;36.4.1 METAL DM Assistant;725
10.3.5;36.5 Future Directions and Conclusions;726
10.3.6;References;727
10.4;37 Bias vs Variance Decomposition For Regression and Classification;731
10.4.1;37.1 Introduction;731
10.4.2;37.2 Bias/Variance Decompositions;733
10.4.2.1;37.2.1 Bias/Variance Decomposition of the Squared Loss;733
10.4.2.2;37.2.2 Bias/variance decompositions of the 0-1 loss;735
10.4.3;37.3 Estimation of Bias and Variance;738
10.4.4;37.4 Experiments and Applications;740
10.4.4.1;37.4.1 Bias/variance tradeoff;740
10.4.4.2;37.4.2 Comparison of some learning algorithms;741
10.4.4.3;37.4.3 Ensemble methods: bagging;742
10.4.5;37.5 Discussion;743
10.4.6;References;743
10.5;38 Mining with Rare Cases;745
10.5.1;38.1 Introduction;745
10.5.2;38.2 Why Rare Cases are Problematic;747
10.5.3;38.3 Techniques for Handling Rare Cases;749
10.5.3.1;38.3.1 Obtain Additional Training Data;749
10.5.3.2;38.3.2 Use a More Appropriate Inductive Bias;750
10.5.3.3;38.3.3 Using More Appropriate Metrics;751
10.5.3.4;38.3.4 Employ Non-Greedy Search Techniques;751
10.5.3.5;38.3.5 Utilize Knowledge/Human Interaction;752
10.5.3.6;38.3.6 Employ Boosting;752
10.5.3.7;38.3.7 Place Rare Cases Into Separate Classes;753
10.5.4;38.4 Conclusion;753
10.5.5;References;754
10.6;39 Data Stream Mining;756
10.6.1;39.1 Introduction;756
10.6.2;39.2 Clustering Techniques;759
10.6.3;39.3 Classification Techniques;761
10.6.4;39.4 Frequent Pattern Mining Techniques;769
10.6.5;39.5 Time Series Analysis;770
10.6.6;39.6 Systems and Applications;771
10.6.7;39.7 Taxonomy of Data Stream Mining Approaches;773
10.6.7.1;39.7.1 Data-based Techniques;773
10.6.7.2;39.7.2 Task-based Techniques;775
10.6.8;39.8 RelatedWork;777
10.6.9;39.9 Future Directions;779
10.6.10;39.10 Summary;779
10.6.11;References;780
10.7;40 Mining Concept-Drifting Data Streams;785
10.7.1;40.1 Introduction;785
10.7.2;40.2 The Data Expiration Problem;787
10.7.3;40.3 Classifier Ensemble for Drifting Concepts;788
10.7.4;40.3.1 Accuracy-Weighted Ensembles;789
10.7.5;40.4 Experiments;791
10.7.6;40.4.1 Algorithms used in Comparison;791
10.7.7;40.4.2 Streaming Data;791
10.7.8;40.4.3 Experimental Results;792
10.7.9;40.5 Discussion and RelatedWork;796
10.7.10;References;797
10.8;41 Mining High-Dimensional Data;799
10.8.1;41.1 Introduction;799
10.8.2;41.2 Chanllenges;800
10.8.3;41.3 Frequent Pa;800
10.8.4;41.4 Clustering;801
10.8.5;41.5 Classification;802
10.8.6;References;803
10.9;42 Text Mining and Information Extraction;805
10.9.1;42.1 Introduction;805
10.9.2;42.2 Text Mining vs. Text Retrieval;807
10.9.3;42.3 Task-Oriented Approaches vs. Formal Frameworks;808
10.9.4;42.4 Task-Oriented Approaches;808
10.9.4.1;42.4.1 Problem Dependant Task - Information Extraction in Text Mining;810
10.9.5;42.5 Formal Frameworks And Algorithm-Based Techniques;812
10.9.5.1;42.5.1 Text Categorization;812
10.9.5.2;42.5.2 Probabilistic models for Information Extraction;815
10.9.6;42.6 Hybrid Approaches - TEG;817
10.9.7;42.7 Text Mining - Visualization and Analytics;818
10.9.7.1;42.7.1 Clear Research;818
10.9.7.2;42.7.2 Other Visualization and Analytical Approaches;822
10.9.8;References;823
10.10;43 Spatial Data Mining;832
10.10.1;43.1 Introduction;832
10.10.2;43.2 Spatial Data;833
10.10.3;43.3 Spatial Outliers;836
10.10.4;43.4 Spatial Co-location Rules;840
10.10.5;43.5 Predictive Models;843
10.10.6;43.6 Spatial Clusters;846
10.10.7;43.7 Summary;847
10.10.8;Acknowledgments;847
10.10.9;References;848
10.11;44 Spatio-temporal clustering;850
10.11.1;44.1 Introduction;850
10.11.2;44.2 Spatio-temporal clustering;851
10.11.2.1;44.2.1 A classification of spatio-temporal data types;851
10.11.2.2;44.2.2 Clustering Methods for Trajectory D;854
10.11.3;44.3 Applications;861
10.11.3.1;44.3.1 Movement data;861
10.11.3.2;44.3.2 Cellular networks;863
10.11.3.3;44.3.3 Environmental data;864
10.11.4;44.4 Open Issues;865
10.11.5;44.5 Conclusions;866
10.11.6;References;866
10.12;45 Data Mining for Imbalanced Datasets: An Overview;870
10.12.1;45.1 Introduction;870
10.12.2;45.2 Performance Measure;871
10.12.2.1;45.2.1 ROC Curves;872
10.12.2.2;45.2.2 Precision and Recall;873
10.12.2.3;45.2.3 Cost-sensitive Measures;874
10.12.3;45.3 Sampling Strategies;874
10.12.3.1;45.3.1 Synthetic Minority Oversampling TEchnique: SMOTE;875
10.12.4;45.4 Ensemble-based Methods;876
10.12.4.1;45.4.1 SMOTEBoost;877
10.12.5;45.5 Discussion;877
10.12.6;Acknowledgements;878
10.12.7;References;878
10.13;46 Relational Data Mining;882
10.13.1;46.1 In a Nutshell;882
10.13.1.1;46.1.1 Relational Data;882
10.13.1.2;46.1.2 Relational Patterns;883
10.13.1.3;46.1.3 Relational to propositional;884
10.13.1.4;46.1.4 Algorithms for relational Data Mining;884
10.13.1.5;46.1.5 Applications of relational Data Mining;885
10.13.1.6;46.1.6 What´s in this chapter;886
10.13.2;46.2 Inductive logic programming;886
10.13.2.1;46.2.1 Logic programs and databases;886
10.13.2.2;46.2.2 The ILP task of relational rule induction;887
10.13.2.3;46.2.3 Structuring the space of clauses;889
10.13.2.4;46.2.4 Searching the space of clauses;890
10.13.2.5;46.2.5 Transforming ILP problems to propositional form;892
10.13.2.6;46.2.6 Upgrading propositional approaches;894
10.13.3;46.3 Relational Association Rules;894
10.13.3.1;46.3.1 Frequent Datalog queries and query extensions;894
10.13.3.2;46.3.2 Discovering frequent queries: WARMR;896
10.13.4;46.4 Relational Decision Trees;898
10.13.4.1;46.4.1 Relational Classification, Regression, and Model Trees;899
10.13.4.2;46.4.2 Induction of Relational Decision Trees;902
10.13.5;46.5 RDM Literature and Internet Resources;903
10.13.6;References;904
10.14;47 Web Mining;907
10.14.1;47.1 Introduction;907
10.14.2;47.2 Graph Properties of theWeb;908
10.14.3;47.3 Web Search;909
10.14.4;47.4 Text Classification;911
10.14.5;47.5 Hypertext Classification;911
10.14.6;47.6 Information Extraction and Wrapper Induction;913
10.14.7;47.7 The SemanticWeb;914
10.14.8;47.8 Web Usage Mining;915
10.14.9;47.9 Collaborative Filtering;915
10.14.10;47.10 Conclusion;916
10.14.11;References;916
10.15;48 A Review of Web Document Clustering Approaches;924
10.15.1;48.1 Introduction;924
10.15.2;48.2 Motivation for Document Clustering;925
10.15.3;48.3 Web Document Clustering Approaches;926
10.15.3.1;48.3.1 Text-based Clustering;927
10.15.3.2;48.3.2 Link-based Clustering;932
10.15.3.3;48.3.3 Hybrid Approaches;934
10.15.4;48.4 Comparison;935
10.15.5;48.5 Conclusions and Open Issues;936
10.15.6;References;936
10.16;49 Causal Discovery;942
10.16.1;49.1 Introduction;942
10.16.2;49.2 Background Knowledge;943
10.16.3;49.3 Theoretical Foundation;945
10.16.4;49.4 Learning a DAG of CN by FDs;946
10.16.4.1;49.4.1 Learning an Ordering of Variables from FDs;946
10.16.4.2;49.4.2 Learning the Markov Boundaries of Undecided Variables;947
10.16.5;49.5 Experimental Results;948
10.16.6;49.6 Conclusion;950
10.16.7;References;950
10.17;50 Ensemble Methods in Supervised Learning;952
10.17.1;50.1 Introduction;952
10.17.2;50.2 Sequential Methodology;953
10.17.2.1;50.2.1 Model-guided Instance Selection;953
10.17.2.2;50.2.2 Incremental Batch Learning;958
10.17.3;50.3 Concurrent Methodology;958
10.17.4;50.4 Combining Classifiers;959
10.17.4.1;50.4.1 Simple Combining Methods;959
10.17.4.2;50.4.2 Meta-combining Methods;962
10.17.5;50.5 Ensemble Diversity;965
10.17.5.1;50.5.1 Manipulating the Inducer;965
10.17.5.2;50.5.2 Manipulating the Training Set;966
10.17.5.3;50.5.3 Measuring the Diversity;966
10.17.6;50.6 Ensemble Size;967
10.17.6.1;50.6.1 Selecting the Ensemble Size;967
10.17.6.2;50.6.2 Pruning Ensembles;967
10.17.7;50.7 Cluster Ensemble;968
10.17.8;References;968
10.18;51 Data Mining using Decomposition Methods;973
10.18.1;51.1 Introduction;973
10.18.2;51.2 Decomposition Advantages;975
10.18.2.1;51.2.1 Increasing Classification Performance (Classification Accuracy);975
10.18.2.2;51.2.2 Scalability to Large Databases;976
10.18.2.3;51.2.3 Increasing Comprehensibility;976
10.18.2.4;51.2.4 Modularity;976
10.18.2.5;51.2.5 Suitability for Parallel Computation;976
10.18.2.6;51.2.6 Flexibility in Techniques Selection;976
10.18.3;51.3 The Elementary Decomposition Methodology;977
10.18.4;51.4 The Decomposer´s Characteristics;981
10.18.4.1;51.4.1 Overview;981
10.18.4.2;51.4.2 The Structure Acquiring Method;981
10.18.4.3;51.4.3 The Mutually Exclusive Property;982
10.18.4.4;51.4.4 The Inducer Usage;983
10.18.4.5;51.4.5 Exhaustiveness;983
10.18.4.6;51.4.6 Combiner Usage;984
10.18.4.7;51.4.7 Sequentially or Concurrently;984
10.18.5;51.5 The Relation to Other Methodologies;985
10.18.6;51.6 Summary;986
10.18.7;References;986
10.19;52 Information Fusion - Methods and Aggregation Operators;991
10.19.1;52.1 Introduction;991
10.19.2;52.2 Preprocessing Data;992
10.19.2.1;52.2.1 Re-identification Algorithms;992
10.19.2.2;52.2.2 Fusion to Improve the Quality of Data;993
10.19.3;52.3 Building Data Models;994
10.19.3.1;52.3.1 Data Models Using Aggregation Operators;995
10.19.3.2;52.3.2 Aggregation Operators to Fuse Data Models;996
10.19.4;52.4 Information Extraction;996
10.19.4.1;52.4.1 Summarization;996
10.19.4.2;52.4.2 Knowledge from Aggregation Operators;997
10.19.5;52.5 Conclusions;997
10.19.6;References;998
10.20;53 Parallel And Grid-Based Data Mining - Algorithms, Models and Systems for High-Performance KDD;1001
10.20.1;53.1 Introduction;1001
10.20.2;53.2 Parallel Data Mining;1003
10.20.2.1;53.2.1 Parallelism in Data Mining Techniques;1003
10.20.2.2;53.2.2 Architectural and Research Issues;1008
10.20.3;53.3 Grid-Based Data Mining;1009
10.20.3.1;53.3.1 Grid-Based Data Mining Systems;1009
10.20.4;53.4 The Knowledge Grid;1013
10.20.4.1;53.4.1 Knowledge Grid Components and Tools;1015
10.20.5;53.5 Summary;1018
10.20.6;References;1018
10.21;54 Collaborative Data Mining;1021
10.21.1;54.1 Introduction;1021
10.21.2;54.2 Remote Collaboration;1022
10.21.2.1;54.2.1 E-Collaboration:Motivations and Forms;1022
10.21.2.2;54.2.2 E-Collaboration Space;1023
10.21.2.3;54.2.3 Collaborative Data Mining in E-Collaboration Space;1023
10.21.3;54.3 The Data Mining Process;1024
10.21.4;54.4 Collaborative Data Mining Guidelines;1025
10.21.4.1;54.4.1 Collaboration Principles;1025
10.21.4.2;54.4.2 Data Mining model evaluation and combination;1026
10.21.5;54.5 Discussion;1028
10.21.6;54.6 Conclusions;1029
10.21.7;References;1029
10.22;55 Organizational Data Mining;1032
10.22.1;55.1 Introduction;1032
10.22.2;55.2 Organizational Data Mining;1033
10.22.3;55.3 ODM versus Data Mining;1034
10.22.3.1;55.3.1 Organizational Theory and ODM;1035
10.22.4;55.4 Ongoing ODM Research;1035
10.22.5;55.5 ODM Advantages;1036
10.22.6;55.6 ODM Evolution;1037
10.22.7;55.6.1 Past;1037
10.22.8;55.6.2 Present;1037
10.22.9;55.6.3 Future;1037
10.22.10;55.7 Summary;1039
10.22.11;References;1039
10.23;56 Mining Time Series Data;1040
10.23.1;56.1 Introduction;1040
10.23.2;56.2 Time Series Similarity Measures;1041
10.23.2.1;56.2.1 Euclidean Distances and Lp Norms;1041
10.23.2.2;56.2.2 Dynamic TimeWarping;1042
10.23.2.3;56.2.3 Longest Common Subsequence Similarity;1043
10.23.2.4;56.2.4 Probabilistic methods;1045
10.23.2.5;56.2.5 General Transformations;1046
10.23.3;56.3 Time Series Data Mining;1046
10.23.3.1;56.3.1 Classification;1047
10.23.3.2;56.3.2 Indexing (Query by Content);1047
10.23.3.3;56.3.3 Clustering;1050
10.23.3.4;56.3.4 Prediction (Forecasting);1051
10.23.3.5;56.3.5 Summarization;1051
10.23.3.6;56.3.6 Anomaly Detection;1054
10.23.3.7;56.3.7 Segmentation;1055
10.23.4;56.4 Time Series Representations;1056
10.23.4.1;56.4.1 Discrete Fourier Transform;1057
10.23.4.2;56.4.2 DiscreteWavelet Transform;1058
10.23.4.3;56.4.3 Singular Value Decomposition;1059
10.23.4.4;56.4.4 Piecewise Linear Approximation;1059
10.23.4.5;56.4.5 Piecewise Aggregate Approximation;1060
10.23.4.6;56.4.6 Adaptive Piecewise Constant Approximation;1060
10.23.4.7;56.4.7 Symbolic Aggregate Approximation (SAX);1062
10.23.5;56.5 Summary;1064
10.23.6;References;1064
11;Part VII Applications;1069
11.1;57 Multimedia Data Mining;1070
11.1.1;57.1 Introduction;1070
11.1.2;57.2 A Typical Architecture of a Multimedia Data Mining System;1074
11.1.3;57.3 An Example- Concept Discovery in Imagery Data;1074
11.1.4;57.3.1 Background and Related Work;1075
11.1.5;57.3.2 Region Based Image Representation;1077
11.1.6;57.3.3 Probabilistic Hidden Semantic Model;1083
11.1.7;57.3.4 Posterior Probability Based Image Mining and Retrieval;1086
11.1.8;57.3.5 Approach Analysis;1088
11.1.9;57.3.6 Experimental Results;1089
11.1.10;57.4 Summary;1093
11.1.11;Ackonwledgments;1094
11.1.12;References;1094
11.2;58 Data Mining in Medicine;1099
11.2.1;58.1 Introduction;1099
11.2.2;58.2 Symbolic Classification Methods;1101
11.2.2.1;58.2.1 Rule Induction;1101
11.2.2.2;58.2.2 Learning of Classification and Regression Trees;1105
11.2.2.3;58.2.3 Inductive Logic Programming;1107
11.2.2.4;58.2.4 Discovery of Concept Hierarchies and Constructive Induction;1107
11.2.2.5;58.2.5 Case-Based Reasoning;1109
11.2.3;58.3 Subsymbolic Classification Methods;1110
11.2.3.1;58.3.1 Instance-Based Learning;1110
11.2.3.2;58.3.2 Neural Networks;1111
11.2.3.3;58.3.3 Bayesian Classifier;1113
11.2.4;58.4 Other Methods Supporting Medical Knowledge Discovery;1114
11.2.5;58.5 Conclusions;1116
11.2.6;Acknowledgments;1117
11.2.7;References;1117
11.3;59 Learning Information Patterns in Biological Databases - Stochastic Data Mining;1125
11.3.1;59.1 Background;1125
11.3.2;59.2 Learning Stochastic Pattern Models;1127
11.3.2.1;59.2.1 Assimilating the Pattern Sets;1127
11.3.2.2;59.2.2 Clustering Biological Patterns;1129
11.3.2.3;59.2.3 Learning Cluster Models;1131
11.3.3;59.3 Searching for Meta-Patterns;1132
11.3.3.1;59.3.1 Level I Search: Locating High Pattern Density Region;1134
11.3.3.2;59.3.2 Level II Search: Meta-Pattern Hypotheses;1137
11.3.4;59.4 Conclusions;1138
11.3.5;References;1139
11.4;60 Data Mining for Financial Applications;1141
11.4.1;60.1 Introduction: Financial Tasks;1141
11.4.2;60.2 Specifics of Data Mining in Finance;1143
11.4.2.1;60.2.1 Time series analysis;1144
11.4.2.2;60.2.2 Data selection and forecast horizon;1144
11.4.2.3;60.2.3 Measures of success;1145
11.4.2.4;60.2.4 QUALITY OF PATTERNS AND HYPOTHESIS EVALUATION;1145
11.4.3;60.3 Aspects of Data Mining Methodology in Finance;1146
11.4.3.1;60.3.1 Attribute-based and relational methodologies;1147
11.4.3.2;60.3.2 Attribute-based relational methodologies;1147
11.4.3.3;60.3.3 Problem ID and method profile;1148
11.4.3.4;60.3.4 Relational Data Mining in finance;1148
11.4.3.5;60.5 Conclusion;1153
11.4.3.6;References;1154
11.4.4;60.4 Data Mining Models and Practice in Finance;1149
11.4.4.1;60.4.1 Portfolio management and neural networks;1149
11.4.4.2;60.4.2 Interpretable trading rules and relational Data Mining;1150
11.4.4.3;60.4.3 Discovering money laundering and attribute-based relational Data Mining;1151
11.4.5;60.5 Conclusion;1153
11.4.6;References;1154
11.5;61 Data Mining for Intrusion Detection;1158
11.5.1;61.1 Introduction;1158
11.5.2;61.2 Data Mining Basics;1159
11.5.3;61.3 Data Mining Meets Intrusion Detection;1161
11.5.3.1;61.3.1 ADAM;1162
11.5.3.2;61.3.2 MADAM ID;1164
11.5.3.3;61.3.3 MINDS;1164
11.5.3.4;61.3.4 Clustering of Unlabeled ID;1165
11.5.3.5;61.3.5 Alert Correlation;1165
11.5.4;61.4 Conclusions and Future Research Directions;1166
11.5.5;References;1166
11.6;62 Data Mining for CRM;1168
11.6.1;62.1 What is CRM?;1168
11.6.2;62.2 Data Mining and Campaign Management;1169
11.6.3;62.3 An Example: Customer Acquisition;1170
11.6.3.1;62.3.1 How Data Mining and Statistical Modeling Changes Things;1171
11.6.3.2;62.3.2 Defining Some Key Acquisition Concepts;1171
11.6.3.3;62.3.3 It All Begins with the Data;1173
11.6.3.4;62.3.4 Test Campaigns;1174
11.6.3.5;62.3.5 Building Data Mining Models Using Response Behaviors;1174
11.7;63 Data Mining for Target Marketing;1176
11.7.1;63.1 Introduction;1176
11.7.2;63.2 Modeling Process;1177
11.7.3;63.3 Evaluation Metrics;1178
11.7.3.1;63.3.1 Gains Charts;1178
11.7.3.2;63.3.2 Prediction Accuracy;1181
11.7.3.3;63.3.3 Profitability/ROI;1181
11.7.3.4;63.3.4 Gains Table;1181
11.7.4;63.4 Segmentation Methods;1182
11.7.4.1;63.4.1 Judgmentally-based RFM/FRAT methods;1182
11.7.4.2;63.4.2 Clustering;1183
11.7.4.3;63.4.3 Classification Methods;1185
11.7.4.4;63.4.4 Decision Making;1186
11.7.5;63.5 Predictive Modeling;1187
11.7.5.1;63.5.1 Linear Regression;1187
11.7.5.2;63.5.2 Logistic Regression;1188
11.7.5.3;63.5.3 Neural Networks;1189
11.7.5.4;63.5.4 Decision Making;1190
11.7.6;63.6 In-Market Timing;1192
11.7.6.1;63.6.1 Logistic Regression;1192
11.7.6.2;63.6.2 Survival Analysis;1193
11.7.7;63.7 Pitfalls of Targeting;1195
11.7.7.1;63.7.1 Modeling Pitfalls;1196
11.7.7.2;63.7.2 Data Pitfalls;1200
11.7.7.3;63.7.3 Implementation Pitfalls;1202
11.7.8;63.8 Conclusions;1205
11.7.8.1;63.8.1 Multiple Offers;1205
11.7.8.2;63.8.2 Multiple Products/Services;1205
11.7.9;References;1206
11.8;64 NHECD - Nano Health and Environmental Commented Database;1208
11.8.1;64.1 Introduction;1208
11.8.2;64.2 The NHECD Model;1216
11.8.3;64.3 NHECD implementation;1217
11.8.3.1;64.3.1 Taxonomies;1217
11.8.3.2;64.3.2 Crawling;1217
11.8.3.3;64.3.3 Information extraction;1220
11.8.3.4;64.3.4 NHECD products;1222
11.8.3.5;64.3.5 Scientific paper rating;1222
11.8.3.6;64.3.6 NHECD Frontend;1223
11.8.4;64.4 Conclusions;1224
11.8.5;64.5 Further research;1226
11.8.6;References;1227
12;Part VIII Software;1229
12.1;65 Commercial Data Mining Software;1230
12.1.1;65.1 Introduction;1230
12.1.2;65.2 Literature Review;1231
12.1.3;65.3 Data Mining Software;1232
12.1.3.1;65.3.1 BioDiscovery GeneSight;1233
12.1.3.2;65.3.2 Megaputer PolyAnalyst 5.0;1233
12.1.3.3;65.3.3 SAS Enterprise Miner;1234
12.1.3.4;65.3.4 PASW Modeler/ Formerly SPSS Clementine;1236
12.1.3.5;65.3.5 IBM DB2 Intelligent Miner;1237
12.1.4;65.4 Supercomputing Data Mining Software;1239
12.1.4.1;65.4.1 Data Visualization using Avizo;1239
12.1.4.2;65.4.2 Data Visualization using JMP Genomics;1241
12.1.5;65.5 Text Mining Software;1243
12.1.5.1;65.5.1 SAS Text Miner;1243
12.1.5.2;65.5.2 Megaputer PolyAnalyst;1245
12.1.6;65.6 Web Mining Software;1246
12.1.6.1;65.6.1 Megaputer PolyAnalyst;1247
12.1.6.2;65.6.2 SPSS Clementine;1249
12.1.7;65.7 Conclusion and Future Research;1250
12.1.8;References;1251
12.2;66 Weka-A Machine LearningWorkbench for Data Mining;1254
12.2.1;66.1 Introduction;1254
12.2.2;Acknowledgments;1261
12.2.3;References;1261
13;Index;1263mehr

Autor

Prof. Oded Maimon is the Oracle chaired Professor at Tel-Aviv University, Previously at MIT. Oded is a leader expert in the field of data mining and knowledge discovery. He published many articles on new algorithms and seven significant award winning books in the field since 2000. He has also developed and implemented successful applications in the Industry. He heads an international research group sponsored by European Union awards.

Dr. Lior Rokach is a senior lecturer at the Department of Information System Engineering at Ben-Gurion University. He is a recognized expert in intelligent information systems and has held several leading positions in this field. His main areas of interest are Data Mining, Pattern Recognition, and Recommender Systems. Dr. Rokach is the author of over 70 refereed papers in leading journals, conference proceedings and book chapters. In addition he has authored six books and edited three others books.