Galit Shmueli, Peter C. Bruce, Nitin R. Patel

Data Mining for Business Analytics

Concepts, Techniques, and Applications with Xlminer

• Gebundenes Buch

Produktbeschreibung

Data Mining for Business Analytics: Concepts, Techniques, and Applications in XLMiner®, Third Edition presents an applied approach to data mining and predictive analytics with clear exposition, hands-on exercises, and real-life case studies. Readers will work with all of the standard data mining methods using the Microsoft® Office Excel® add-in XLMiner® to develop predictive models and learn how to obtain business value from Big Data.Featuring updated topical coverage on text mining, social network analysis, collaborative filtering, ensemble methods, uplift modeling and more, the Third Edition also includes:_ Real-world examples to build a theoretical and practical understanding of key data mining methods_ End-of-chapter exercises that help readers better understand the presented material_ Data-rich case studies to illustrate various applications of data mining techniques_ Completely new chapters on social network analysis and text mining_ A companion site with additional data sets, instructors material that include solutions to exercises and case studies, and Microsoft PowerPoint® slides https://www.dataminingbook.com_ Free 140-day license to use XLMiner for Education softwareData Mining for Business Analytics: Concepts, Techniques, and Applications in XLMiner®, Third Edition is an ideal textbook for upper-undergraduate and graduate-level courses as well as professional programs on data mining, predictive modeling, and Big Data analytics. The new edition is also a unique reference for analysts, researchers, and practitioners working with predictive analytics in the fields of business, finance, marketing, computer science, and information technology.Praise for the Second Edition"...full of vivid and thought-provoking anecdotes... needs to be read by anyone with a serious interest in research and marketing."- Research Magazine"Shmueli et al. have done a wonderful job in presenting the field of data mining - a welcome addition to the literature." - ComputingReviews.com"Excellent choice for business analysts...The book is a perfect fit for its intended audience." - Keith McCormick, Consultant and Author of SPSS Statistics For Dummies, Third Edition and SPSS Statistics for Data Analysis and VisualizationGalit 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, The 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 journal articles, books, textbooks and book chapters.Peter C. Bruce is President and Founder of the Institute for Statistics Education at www.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, also published by Wiley.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.

Produktdetails

• Verlag: Wiley / Wiley & Sons
• Artikelnr. des Verlages: 1W118729270
• 3. Aufl.
• Seitenzahl: 560
• Erscheinungstermin: 18. April 2016
• Englisch
• Abmessung: 254mm x 566mm x 33mm
• Gewicht: 666g
• ISBN-13: 9781118729274
• ISBN-10: 1118729277
• Artikelnr.: 45295461

Autorenporträt

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, The 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 journal articles, books, textbooks and book chapters. Peter C. Bruce is President and Founder of the Institute for Statistics Education at www.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, also published by Wiley. 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.

Inhaltsangabe

Foreword xviiPreface to the Third Edition xixPreface to the First Edition xxiiAcknowledgments xxivPART I PRELIMINARIESCHAPTER 1 Introduction 31.1 What is Business Analytics? 31.2 What is Data Mining? 51.3 Data Mining and Related Terms 51.4 Big Data 61.5 Data Science 71.6 Why Are There So Many Different Methods? 81.7 Terminology and Notation 91.8 Road Maps to This Book 11Order of Topics 12CHAPTER 2 Overview of the Data Mining Process 142.1 Introduction 142.2 Core Ideas in Data Mining 152.3 The Steps in Data Mining 182.4 Preliminary Steps 202.5 Predictive Power and Overfitting 262.6 Building a Predictive Model with XLMiner 302.7 Using Excel for Data Mining 402.8 Automating Data Mining Solutions 40Data Mining Software Tools (by Herb Edelstein) 42Problems 45PART II DATA EXPLORATION AND DIMENSION REDUCTIONCHAPTER 3 Data Visualization 503.1 Uses of Data Visualization 503.2 Data Examples 52Example 1: Boston Housing Data 52Example 2: Ridership on Amtrak Trains 533.3 Basic Charts: Bar Charts, Line Graphs, and Scatter Plots 53Distribution Plots 54Heatmaps: Visualizing Correlations and Missing Values 573.4 Multi-Dimensional Visualization 58Adding Variables 59Manipulations 61Reference: trend line and labels 64Scaling up to Large Datasets 65Multivariate Plot 66Interactive Visualization 673.5 Specialized Visualizations 70Visualizing Networked Data 70Visualizing Hierarchical Data: Treemaps 72Visualizing Geographical Data: Map Charts 733.6 Summary: Major Visualizations and Operations, by Data Mining Goal 75Prediction 75Classification 75Time Series Forecasting 75Unsupervised Learning 76Problems 77CHAPTER 4 Dimension Reduction 794.1 Introduction 794.2 Curse of Dimensionality 804.3 Practical Considerations 80Example 1: House Prices in Boston 804.4 Data Summaries 814.5 Correlation Analysis 844.6 Reducing the Number of Categories in Categorical Variables 854.7 Converting A Categorical Variable to A Numerical Variable 864.8 Principal Components Analysis 86Example 2: Breakfast Cereals 87Principal Components 92Normalizing the Data 93Using Principal Components for Classification and Prediction 944.9 Dimension Reduction Using Regression Models 964.10 Dimension Reduction Using Classification and Regression Trees 96Problems 97PART III PERFORMANCE EVALUATIONCHAPTER 5 Evaluating Predictive Performance 1015.1 Introduction 1015.2 Evaluating Predictive Performance 102Benchmark: The Average 102Prediction Accuracy Measures 1035.3 Judging Classifier Performance 106Benchmark: The Naive Rule 107Class Separation 107The Classification Matrix 107Using the Validation Data 109Accuracy Measures 109Cutoff for Classification 110Performance in Unequal Importance of Classes 114Asymmetric Misclassification Costs 1165.4 Judging Ranking Performance 1195.5 Oversampling 123Problems 129PART IV PREDICTION AND CLASSIFICATION METHODSCHAPTER 6 Multiple Linear Regression 1346.1 Introduction 1346.2 Explanatory vs. Predictive Modeling 1356.3 Estimating the Regression Equation and Prediction 136Example: Predicting the Price of Used Toyota Corolla Cars 1376.4 Variable Selection in Linear Regression 141Reducing the Number of Predictors 141How to Reduce the Number of Predictors 142Problems 147CHAPTER 7 k-Nearest Neighbors (kNN) 1517.1 The k-NN Classifier (categorical outcome) 151Determining Neighbors 151Classification Rule 152Example: Riding Mowers 152Choosing k 154Setting the Cutoff Value 1547.2 k-NN for a Numerical Response 1567.3 Advantages and Shortcomings of k-NN Algorithms 158Problems 160CHAPTER 8 The Naive Bayes Classifier 1628.1 Introduction 162Example 1: Predicting Fraudulent Financial Reporting 1638.2 Applying the Full (Exact) Bayesian Classifier 1648.3 Advantages and Shortcomings of the Naive Bayes Classifier 172Advantages and Shortcomings of the naive Bayes Classifier 172Problems 176CHAPTER 9 Classification and Regression Trees 1789.1 Introduction 1789.2 Classification Trees 179Example 1: Riding Mowers 1809.3 Measures of Impurity 1839.4 Evaluating the Performance of a Classification Tree 187Example 2: Acceptance of Personal Loan 1889.5 Avoiding Overfitting 192Stopping Tree Growth: CHAID 192Pruning the Tree 1939.6 Classification Rules from Trees 1989.7 Classification Trees for More Than two Classes 1989.8 Regression Trees 198Prediction 199Measuring Impurity 200Evaluating Performance 2009.9 Advantages and Weaknesses of a Tree 2009.10 Improving Prediction: Multiple Trees 202Problems 205CHAPTER 10 Logistic Regression 20910.1 Introduction 20910.2 The Logistic Regression Model 211Example: Acceptance of Personal Loan 212Model with a Single Predictor 214Estimating the Logistic Model from Data 215Interpreting Results in Terms of Odds 21810.3 Evaluating Classification Performance 219Variable Selection 22010.4 Example of Complete Analysis: Predicting Delayed Flights 222Data Preprocessing 224Model Fitting and Estimation 224Model Interpretation 226Model Performance 226Variable Selection 22710.5 Appendix: Logistic Regression for Profiling 231Appendix A: Why Linear Regression Is Problematic for a Categorical Response 231Appendix B: Evaluating Explanatory Power 233Appendix C: Logistic Regression for More Than Two Classes 235Problems 239CHAPTER 11 Neural Nets 24211.1 Introduction 24211.2 Concept and Structure of a Neural Network 24311.3 Fitting a Network to Data 243Example 1: Tiny Dataset 244Computing Output of Nodes 245Preprocessing the Data 248Training the Model 248Example 2: Classifying Accident Severity 253Avoiding overfitting 254Using the Output for Prediction and Classification 25811.4 Required User Input 25811.5 Exploring the Relationship Between Predictors and Response 25911.6 Advantages and Weaknesses of Neural Networks 261Problems 262CHAPTER 12 Discriminant Analysis 26412.1 Introduction 264Example 1: Riding Mowers 265Example 2: Personal Loan Acceptance 26512.2 Distance of an Observation from a Class 26712.3 Fisher's Linear Classification Functions 26812.4 Classification Performance of Discriminant Analysis 27212.5 Prior Probabilities 27312.6 Unequal Misclassification Costs 27412.7 Classifying More Than Two Classes 274Example 3: Medical Dispatch to Accident Scenes 27412.8 Advantages and Weaknesses 277Problems 279CHAPTER 13 Combining Methods: Ensembles and Uplift Modeling 28213.1 Ensembles 282Why Ensembles Can Improve Predictive Power 283Simple Averaging 284Bagging 286Boosting 286Advantages and Weaknesses of Ensembles 28613.2 Uplift (Persuasion) Modeling 287A-B Testing 287Uplift 288Gathering the Data 288A Simple Model 289Modeling Individual Uplift 290Using the Results of an Uplift Model 29213.3 Summary 292Problems 293PART V MINING RELATIONSHIPS AMONG RECORDSCHAPTER 14 Association Rules and Collaborative Filtering 29714.1 Association Rules 297Discovering Association Rules in Transaction Databases 298Example 1: Purchases of Phone Faceplates 298Generating Candidate Rules 298The Apriori Algorithm 301Selecting Strong Rules 301Data Format 303The Process of Rule Selection 304Interpreting the Results 306Rules and Chance 306Example 2: Rules for Similar Book Purchases 30814.2 Collaborative Filtering1 310Data Type and Format 311Example 3: Netflix Prize Contest 311User-Based Collaborative Filtering: "People Like You" 312Item-Based Collaborative Filtering 315Advantages and Weaknesses of Collaborative Filtering 316Collaborative Filtering vs. Association Rules 31614.3 Summary 318Problems 320CHAPTER 15 Cluster Analysis 32415.1 Introduction 324Example: Public Utilities 32615.2 Measuring Distance Between Two Observations 328Euclidean Distance 328Normalizing Numerical Measurements 328Other Distance Measures for Numerical Data 329Distance Measures for Categorical Data 331Distance Measures for Mixed Data 33115.3 Measuring Distance Between Two Clusters 33215.4 Hierarchical (Agglomerative) Clustering 334Single Linkage 335Complete Linkage 335Average Linkage 336Centroid Linkage 336Dendrograms: Displaying Clustering Process and Results 337Validating Clusters 339Limitations of Hierarchical Clustering 34015.5 Non-hierarchical Clustering: The k-Means Algorithm 341Initial Partition into k Clusters 342Problems 346PART VI FORECASTING TIME SERIESCHAPTER 16 Handling Time Series 35116.1 Introduction 35116.2 Descriptive vs. Predictive Modeling 35216.3 Popular Forecasting Methods in Business 353Combining Methods 35316.4 Time Series Components 354Example: Ridership on Amtrak Trains 35416.5 Data Partitioning and Performance Evaluation 358Benchmark Performance: Naive Forecasts 359Generating Future Forecasts 359Problems 361CHAPTER 17 Regression-Based Forecasting 36417.1 A Model with Trend 364Linear Trend 364Exponential Trend 366Polynomial Trend 36917.2 A Model with Seasonality 37017.3 A model with trend and seasonality 37117.4 Autocorrelation and ARIMA Models 371Computing Autocorrelation 374Improving Forecasts by Integrating Autocorrelation Information 376Evaluating Predictability 380Problems 382CHAPTER 18 Smoothing Methods 39218.1 Introduction 39218.2 Moving Average 393Centered Moving Average for Visualization 393Trailing Moving Average for Forecasting 395Choosing Window Width (w) 39918.3 Simple Exponential Smoothing 399Choosing Smoothing Parameter 400Relation Between Moving Average and Simple Exponential Smoothing 40118.4 Advanced Exponential Smoothing 402Series with a Trend 402Series with a Trend and Seasonality 403Series with Seasonality (No Trend) 403Problems 405PART VII DATA ANALYTICSCHAPTER 19 Social Network Analytics 41519.1 Introduction 41519.2 Directed vs. Undirected Networks 41619.3 Visualizing and analyzing networks 418Graph Layout 418Adjacency List 421Adjacency Matrix 422Using Network Data in Classification and Prediction 42219.4 Social Data Metrics and Taxonomy 423Node-Level Centrality Metrics 423Egocentric Network 424Network Metrics 42519.5 Using Network Metrics in Prediction and Classification 427Link Prediction 427Entity Resolution 427Collaborative Filtering 428Advantages and Disadvantages 431Problems 434CHAPTER 20 Text Mining 43620.1 Introduction 43620.2 The Spreadsheet Representation of Text: "Bag-of-Words" 43720.3 Bag-of-Words vs. Meaning Extraction at Document Level 43720.4 Preprocessing the Text 438Tokenization 439Text Reduction 439Presence/Absence vs. Frequency 440Term Frequency - Inverse Document Frequency (TF-IDF) 441From Terms to Concepts: Latent Semantic Indexing 441Extracting Meaning 44120.5 Implementing data mining methods 44220.6 Example: Online Discussions on Autos and Electronics 442Importing and Labeling the Records 443Tokenization 444Text Processing and Reduction 444Producing a Concept Matrix 444Labeling the Documents 447Fitting a Model 447Prediction 44920.7 Summary 449Problems 450PART VIII CASESCHAPTER 21 Cases 45421.1 Charles Book Club2 45421.2 German Credit 46321.3 Tayko Software Cataloger3 46821.4 Political Persuasion4 47221.5 Taxi Cancellations5 47521.6 Segmenting Consumers of Bath Soap6 47721.7 Direct-Mail Fundraising 48021.8 Catalog Cross-Selling7 48321.9 Predicting Bankruptcy 48421.10Time Series Case: Forecasting Public Transportation Demand 487References 489