WOUTER VERBEKE is assistant professor of Business Informatics and Data Analytics at Vrije Universiteit Brussel (Belgium). He is the coauthor of Fraud Analytics using Descriptive, Predictive, and Social Network Techniques.BART BAESENS is a professor at KU Leuven (Belgium) and a lecturer at the University of Southampton (United Kingdom). He is the author of Credit Risk Management and Analytics in a Big Data World, as well as coauthor of Fraud Analytics using Descriptive, Predictive, and Social Network Techniques. CRISTIÁN BRAVO is a lecturer vin business analytics in the department of Decision Analytics and Risk at the University of Southampton.

• Foreword xvAcknowledgments xviiChapter 1 A Value-Centric Perspective Towards Analytics 1Introduction 1Business Analytics 3Profit-Driven Business Analytics 9Analytics Process Model 14Analytical Model Evaluation 17Analytics Team 19Profiles 19Data Scientists 20Conclusion 23Review Questions 24Multiple Choice Questions 24Open Questions 25References 25Chapter 2 Analytical Techniques 28Introduction 28Data Preprocessing 29Denormalizing Data for Analysis 29Sampling 30Exploratory Analysis 31Missing Values 31Outlier Detection and Handling 32Principal Component Analysis 33Types of Analytics 37Predictive Analytics 37Introduction 37Linear Regression 38Logistic Regression 39Decision Trees 45Neural Networks 52Ensemble Methods 56Bagging 57Boosting 57Random Forests 58Evaluating Ensemble Methods 59Evaluating Predictive Models 59Splitting Up the Dataset 59Performance Measures for Classification Models 63Performance Measures for Regression Models 67Other Performance Measures for Predictive AnalyticalModels 68Descriptive Analytics 69Introduction 69Association Rules 69Sequence Rules 72Clustering 74Survival Analysis 81Introduction 81Survival Analysis Measurements 83Kaplan Meier Analysis 85Parametric Survival Analysis 87Proportional Hazards Regression 90Extensions of Survival Analysis Models 92Evaluating Survival Analysis Models 93Social Network Analytics 93Introduction 93Social Network Definitions 94Social Network Metrics 95Social Network Learning 97Relational Neighbor Classifier 98Probabilistic Relational Neighbor Classifier 99Relational Logistic Regression 100Collective Inferencing 102Conclusion 102Review Questions 103Multiple Choice Questions 103Open Questions 108Notes 110References 110Chapter 3 Business Applications 114Introduction 114Marketing Analytics 114Introduction 114RFM Analysis 115Response Modeling 116Churn Prediction 118X-selling 120Customer Segmentation 121Customer Lifetime Value 123Customer Journey 129Recommender Systems 131Fraud Analytics 134Credit Risk Analytics 139HR Analytics 141Conclusion 146Review Questions 146Multiple Choice Questions 146Open Questions 150Note 151References 151Chapter 4 Uplift Modeling 154Introduction 154The Case for Uplift Modeling: Response Modeling 155Effects of a Treatment 158Experimental Design, Data Collection, and DataPreprocessing 161Experimental Design 161Campaign Measurement of Model Effectiveness 164Uplift Modeling Methods 170Two-Model Approach 172Regression-Based Approaches 174Tree-Based Approaches 183Ensembles 193Continuous or Ordered Outcomes 198Evaluation of Uplift Models 199Visual Evaluation Approaches 200Performance Metrics 207Practical Guidelines 210Two-Step Approach for Developing Uplift Models 210Implementations and Software 212Conclusion 213Review Questions 214Multiple Choice Questions 214Open Questions 216Note 217References 217Chapter 5 Profit-Driven Analytical Techniques 220Introduction 220Profit-Driven Predictive Analytics 221The Case for Profit-Driven Predictive Analytics 221Cost Matrix 222Cost-Sensitive Decision Making with Cost-InsensitiveClassification Models 228Cost-Sensitive Classification Framework 231Cost-Sensitive Classification 234Pre-Training Methods 235During-Training Methods 247Post-Training Methods 253Evaluation of Cost-Sensitive Classification Models 255Imbalanced Class Distribution 256Implementations 259Cost-Sensitive Regression 259The Case for Profit-Driven Regression 259Cost-Sensitive Learning for Regression 260During Training Methods 260Post-Training Methods 261Profit-Driven Descriptive Analytics 267Profit-Driven Segmentation 267Profit-Driven Association Rules 280Conclusion 283Review Questions 284Multiple Choice Questions 284Open Questions 289Notes 290References 291Chapter 6 Profit-Driven Model Evaluationand Implementation 296Introduction 296Profit-Driven Evaluation of Classification Models 298Average Misclassification Cost 298Cutoff Point Tuning 303ROC Curve-Based Measures 310Profit-Driven Evaluation with Observation-DependentCosts 334Profit-Driven Evaluation of Regression Models 338Loss Functions and Error-Based Evaluation Measures 339REC Curve and Surface 341Conclusion 345Review Questions 347Multiple Choice Questions 347Open Questions 350Notes 351References 352Chapter 7 Economic Impact 355Introduction 355Economic Value of Big Data and Analytics 355Total Cost of Ownership (TCO) 355Return on Investment (ROI) 357Profit-Driven Business Analytics 359Key Economic Considerations 359In-Sourcing versus Outsourcing 359On Premise versus the Cloud 361Open-Source versus Commercial Software 362Improving the ROI of Big Data and Analytics 364New Sources of Data 364Data Quality 367Management Support 369Organizational Aspects 370Cross-Fertilization 371Conclusion 372Review Questions 373Multiple Choice Questions 373Open Questions 376Notes 377References 377About the Authors 378Index 381