Data Mining for Business Analytics

Data Mining for Business Analytics: Concepts, Techniques, and Applications in R presents an applied approach to data mining concepts and methods, using R software for illustration

Readers will learn how to implement a variety of popular data mining algorithms in R (a free and open-source software) to tackle business problems and opportunities.

This is the fifth version of this successful text, and the first using R. It covers both statistical and machine learning algorithms for prediction, classification, visualization, dimension reduction, recommender systems, clustering, text mining and network analysis. It also includes:

• Two new co-authors, Inbal Yahav and Casey Lichtendahl, who bring both expertise teaching business analytics courses using R, and data mining consulting experience in business and government
• Updates and new material based on feedback from instructors teaching MBA, undergraduate, diploma and executive courses, and from their students
• More than a dozen case studies demonstrating applications for the data mining techniques described
• End-of-chapter exercises that help readers gauge and expand their comprehension and competency of the material presented
• A companion website with more than two dozen data sets, and instructor materials including exercise solutions, PowerPoint slides, and case solutions  www.dataminingbook.com

Data Mining for Business Analytics: Concepts, Techniques, and Applications in R is an ideal textbook for graduate and upper-undergraduate level courses in data mining, predictive analytics, and business analytics. This new edition is also an excellent reference for analysts, researchers, and practitioners working with quantitative methods in the fields of business, finance, marketing, computer science, and information technology.

Galit Shmueli, PhD, is Distinguished Professor at National Tsing Hua University's Institute of Service Science. She has designed and instructed data mining courses since 2004 at University of Maryland, Statistics.com, Indian School of Business, and National Tsing Hua University, Taiwan. Professor Shmueli is known for her research and teaching in business analytics, with a focus on statistical and data mining methods in information systems and healthcare. She has authored over 70 publications including books.

Peter C. Bruce is President and Founder of the Institute for Statistics Education at Statistics.com. He has written multiple journal articles and is the developer of Resampling Stats software. He is the author of Introductory Statistics and Analytics: A Resampling Perspective (Wiley) and co-author of Practical Statistics for Data Scientists: 50 Essential Concepts (O'Reilly).

Inbal Yahav, PhD, is Professor at the Graduate School of Business Administration at Bar-Ilan University, Israel. She teaches courses in social network analysis, advanced research methods, and software quality assurance. Dr. Yahav received her PhD in Operations Research and Data Mining from the University of Maryland, College Park.

Nitin R. Patel, PhD, is Chairman and cofounder of Cytel, Inc., based in Cambridge, Massachusetts. A Fellow of the American Statistical Association, Dr. Patel has also served as a Visiting Professor at the Massachusetts Institute of Technology and at Harvard University. He is a Fellow of the Computer Society of India and was a professor at the Indian Institute of Management, Ahmedabad, for 15 years.

Kenneth C. Lichtendahl, Jr., PhD, is Associate Professor at the University of Virginia. He is the Eleanor F. and Phillip G. Rust Professor of Business Administration and teaches MBA courses in decision analysis, data analysis and optimization, and managerial quantitative analysis. He also teaches executive education courses in strategic analysis and decision-making, and managing the corporate aviation function.

Foreword by Gareth James xix

Foreword by Ravi Bapna xxi

Preface to the R Edition xxiii

Acknowledgments xxvii

PART I PRELIMINARIES

CHAPTER 1 Introduction 3

1.1 What Is Business Analytics? 3

1.2 What Is Data Mining? 5

1.3 Data Mining and Related Terms 5

1.4 Big Data 6

1.5 Data Science 7

1.6 Why Are There So Many Different Methods? 8

1.7 Terminology and Notation 9

1.8 Road Maps to This Book 11

Order of Topics 11

CHAPTER 2 Overview of the Data Mining Process 15

2.1 Introduction 15

2.2 Core Ideas in Data Mining 16

2.3 The Steps in Data Mining 19

2.4 Preliminary Steps 21

2.5 Predictive Power and Overfitting 33

2.6 Building a Predictive Model 38

2.7 Using R for Data Mining on a Local Machine 43

2.8 Automating Data Mining Solutions 43

PART II DATA EXPLORATION AND DIMENSION REDUCTION

CHAPTER 3 Data Visualization 55

3.1 Uses of Data Visualization 55

3.2 Data Examples 57

3.3 Basic Charts: Bar Charts, Line Graphs, and Scatter Plots 59

3.4 Multidimensional Visualization 67

3.5 Specialized Visualizations 80

3.6 Summary: Major Visualizations and Operations, by Data Mining Goal 86

CHAPTER 4 Dimension Reduction 91

4.1 Introduction 91

4.2 Curse of Dimensionality 92

4.3 Practical Considerations 92

4.4 Data Summaries 94

4.5 Correlation Analysis 97

4.6 Reducing the Number of Categories in Categorical Variables 99

4.7 Converting a Categorical Variable to a Numerical Variable 99

4.8 Principal Components Analysis 101

4.9 Dimension Reduction Using Regression Models 111

4.10 Dimension Reduction Using Classification and Regression Trees 111

PART III PERFORMANCE EVALUATION

CHAPTER 5 Evaluating Predictive Performance 117

5.1 Introduction 117

5.2 Evaluating Predictive Performance 118

5.3 Judging Classifier Performance 122

5.4 Judging Ranking Performance 136

5.5 Oversampling 140

PART IV PREDICTION AND CLASSIFICATION METHODS

CHAPTER 6 Multiple Linear Regression 153

6.1 Introduction 153

6.2 Explanatory vs. Predictive Modeling 154

6.3 Estimating the Regression Equation and Prediction 156

6.4 Variable Selection in Linear Regression 161

CHAPTER 7 k-Nearest Neighbors (kNN) 173

7.1 The k-NN Classifier (Categorical Outcome) 173

7.2 k-NN for a Numerical Outcome 180

7.3 Advantages and Shortcomings of k-NN Algorithms 182

CHAPTER 8 The Naive Bayes Classifier 187

8.1 Introduction 187

8.2 Applying the Full (Exact) Bayesian Classifier 189

8.3 Advantages and Shortcomings of the Naive Bayes Classifier 199

CHAPTER 9 Classification and Regression Trees 205

9.1 Introduction 205

9.2 Classification Trees 207

9.3 Evaluating the Performance of a Classification Tree 215

9.4 Avoiding Overfitting 216

9.5 Classification Rules from Trees 226

9.6 Classification Trees for More Than Two Classes 227

9.7 Regression Trees 227

9.8 Improving Prediction: Random Forests and Boosted Trees 229

9.9 Advantages and Weaknesses of a Tree 232

CHAPTER 10 Logistic Regression 237

10.1 Introduction 237

10.2 The Logistic Regression Model 239

10.3 Example: Acceptance of Personal Loan 240

10.4 Evaluating Classification Performance 247

10.5 Example of Complete Analysis: Predicting Delayed Flights 250

10.6 Appendix: Logistic Regression for Profiling 259

Appendix A: Why Linear Regression Is Problematic for a Categorical Outcome 259

Appendix B: Evaluating Explanatory Power 261

Appendix C: Logistic Regression for More Than Two Classes 264

CHAPTER 11 Neural Nets 271

11.1 Introduction 271

11.2 Concept and Structure of a Neural Network 272

11.3 Fitting a Network to Data 273

11.4 Required User Input 285

11.5 Exploring the Relationship between Predictors and Outcome 287

11.6 Advantages and Weaknesses of Neural Networks 288

CHAPTER 12 Discriminant Analysis 293

12.1 Introduction 293

12.2 Distance of a Record from a Class 296

12.3 Fisher’s Linear Classification Functions 297

12.4 Classification Performance of Discriminant Analysis 300

12.5 Prior Probabilities 302

12.6 Unequal Misclassification Costs 302

12.7 Classifying More Than Two Classes 303

12.8 Advantages and Weaknesses 306

CHAPTER 13 Combining Methods: Ensembles and Uplift Modeling 311

13.1 Ensembles 311

13.2 Uplift (Persuasion) Modeling 317

13.3 Summary 324

PART V MINING RELATIONSHIPS AMONG RECORDS

CHAPTER 14 Association Rules and Collaborative Filtering 329

14.1 Association Rules 329

14.2 Collaborative Filtering 342

14.3 Summary 351

CHAPTER 15 Cluster Analysis 357

15.1 Introduction 357

15.2 Measuring Distance between Two Records 361

15.3 Measuring Distance between Two Clusters 366

15.4 Hierarchical (Agglomerative) Clustering 368

15.5 Non-Hierarchical Clustering: The k-Means Algorithm 376

PART VI FORECASTING TIME SERIES

CHAPTER 16 Handling Time Series 387

16.1 Introduction 387

16.2 Descriptive vs. Predictive Modeling 389

16.3 Popular Forecasting Methods in Business 389

16.4 Time Series Components 390

16.5 Data-Partitioning and Performance Evaluation 395

CHAPTER 17 Regression-Based Forecasting 401

17.1 A Model with Trend 401

17.2 A Model with Seasonality 407

17.3 A Model with Trend and Seasonality 411

17.4 Autocorrelation and ARIMA Models 412

CHAPTER 18 Smoothing Methods 433

18.1 Introduction 433

18.2 Moving Average 434

18.3 Simple Exponential Smoothing 439

18.4 Advanced Exponential Smoothing 442

PART VII DATA ANALYTICS

CHAPTER 19 Social Network Analytics 455

19.1 Introduction 455

19.2 Directed vs. Undirected Networks 457

19.3 Visualizing and Analyzing Networks 458

19.4 Social Data Metrics and Taxonomy 462

19.5 Using Network Metrics in Prediction and Classification 467

19.6 Collecting Social Network Data with R 471

19.7 Advantages and Disadvantages 474

CHAPTER 20 Text Mining 479

20.1 Introduction 479

20.2 The Tabular Representation of Text: Term-Document Matrix and “Bag-of-Words” 480

20.3 Bag-of-Words vs. Meaning Extraction at Document Level 481

20.4 Preprocessing the Text 482

20.5 Implementing Data Mining Methods 489

20.6 Example: Online Discussions on Autos and Electronics 490

20.7 Summary 494

PART VIII CASES

CHAPTER 21 Cases 499

21.1 Charles Book Club 499

21.2 German Credit 505

21.3 Tayko Software Cataloger 510

21.4 Political Persuasion 513

21.5 Taxi Cancellations 517

21.6 Segmenting Consumers of Bath Soap 518

21.7 Direct-Mail Fundraising 521

21.8 Catalog Cross-Selling 524

21.9 Predicting Bankruptcy 525

21.10 Time Series Case: Forecasting Public Transportation Demand 528

Index 535