Front Cover -- Data Mining Applications with R -- Copyright -- Contents -- Preface -- Background -- Objectives and Significance -- Target Audience -- Acknowledgments -- Review Committee -- Additional Reviewers -- Foreword -- References -- Chapter 1: Power Grid Data Analysis with R and Hadoop -- 1.1. Introduction -- 1.2. A Brief Overview of the Power Grid -- 1.3. Introduction to MapReduce, Hadoop, and RHIPE -- 1.3.1. MapReduce -- 1.3.1.1. An Example: The Iris Data -- 1.3.2. Hadoop -- 1.3.3. RHIPE: R with Hadoop -- 1.3.3.1. Installation -- 1.3.3.2. Iris MapReduce Example with RHIPE -- 1.3.3.2.1. The Map Expression -- 1.3.3.2.2. The Reduce Expression -- 1.3.3.2.3. Running the Job -- 1.3.3.2.4. Looking at Results -- 1.3.4. Other Parallel R Packages -- 1.4. Power Grid Analytical Approach -- 1.4.1. Data Preparation -- 1.4.2. Exploratory Analysis and Data Cleaning -- 1.4.2.1. 5-min Summaries -- 1.4.2.2. Quantile Plots of Frequency -- 1.4.2.3. Tabulating Frequency by Flag -- 1.4.2.4. Distribution of Repeated Values -- 1.4.2.5. White Noise -- 1.4.3. Event Extraction -- 1.4.3.1. OOS Frequency Events -- 1.4.3.2. Finding Generator Trip Features -- 1.4.3.3. Creating Overlapping Frequency Data -- 1.5. Discussion and Conclusions -- Appendix -- References -- Chapter 2: Picturing Bayesian Classifiers: A Visual Data Mining Approach to Parameters Optimization -- 2.1. Introduction -- 2.2. Related Works -- 2.3. Motivations and Requirements -- 2.3.1. R Packages Requirements -- 2.4. Probabilistic Framework of NB Classifiers -- 2.4.1. Choosing the Model -- 2.4.1.1. Multivariate Bernoulli model -- 2.4.1.2. Multinomial Model -- 2.4.1.3. Poisson Model -- 2.4.2. Estimating the Parameters -- 2.5. Two-Dimensional Visualization System -- 2.5.1. Design Choices -- 2.5.2. Visualization Design -- 2.6. A Case Study: Text Classification -- 2.6.1. Description of the Dataset.

2.6.2. Creating Document-Term Matrices -- 2.6.3. Loading Existing Term-Document Matrices -- 2.6.4. Running the Program -- 2.6.4.1. Comparing Models -- 2.7. Conclusions -- Acknowledgments -- References -- Chapter 3: Discovery of Emergent Issues and Controversies in Anthropology Using Text Mining, Topic Modeling, and Social Ne ... -- 3.1. Introduction -- 3.2. How Many Messages and How Many Twitter-Users in the Sample? -- 3.3. Who Is Writing All These Twitter Messages? -- 3.4. Who Are the Influential Twitter-Users in This Sample? -- 3.5. What Is the Community Structure of These Twitter-Users? -- 3.6. What Were Twitter-Users Writing About During the Meeting? -- 3.7. What Do the Twitter Messages Reveal About the Opinions of Their Authors? -- 3.8. What Can Be Discovered in the Less Frequently Used Words in the Sample? -- 3.9. What Are the Topics That Can Be Algorithmically Discovered in This Sample? -- 3.10. Conclusion -- References -- Chapter 4: Text Mining and Network Analysis of Digital Libraries in R -- 4.1. Introduction -- 4.2. Dataset Preparation -- 4.3. Manipulating the Document-Term Matrix -- 4.3.1. The Document-Term Matrix -- 4.3.2. Term Frequency-Inverse Document Frequency -- 4.3.3. Exploring the Document-Term Matrix -- 4.4. Clustering Content by Topics Using the LDA -- 4.4.1. The Latent Dirichlet Allocation -- 4.4.2. Learning the Various Distributions for LDA -- 4.4.3. Using the Log-Likelihood for Model Validation -- 4.4.4. Topics Representation -- 4.4.5. Plotting the Topics Associations -- 4.5. Using Similarity Between Documents to Explore Document Cohesion -- 4.5.1. Computing Similarities Between Documents -- 4.5.2. Using a Heatmap to Illustrate Clusters of Documents -- 4.6. Social Network Analysis of Authors -- 4.6.1. Constructing the Network as a Graph -- 4.6.2. Author Importance Using Centrality Measures -- 4.7. Conclusion -- References.

Chapter 5: Recommender Systems in R -- 5.1. Introduction -- 5.2. Business Case -- 5.3. Evaluation -- 5.4. Collaborative Filtering Methods -- 5.5. Latent Factor Collaborative Filtering -- 5.6. Simplified Approach -- 5.7. Roll Your Own -- 5.8. Final Thoughts -- References -- Chapter 6: Response Modeling in Direct Marketing: A Data Mining-Based Approach for Target Selection -- 6.1. Introduction/Background -- 6.2. Business Problem -- 6.3. Proposed Response Model -- 6.4. Modeling Detail -- 6.4.1. Data Collection -- 6.4.2. Data Preprocessing -- 6.4.2.1. Data Integration and Cleaning -- 6.4.2.2. Data Normalization -- 6.4.3. Feature Construction -- 6.4.3.1. Target Variable Construction -- 6.4.3.2. Predictor Variables -- 6.4.3.3. Interaction Variables -- 6.4.4. Feature Selection -- 6.4.4.1. F-Score -- 6.4.4.2. Step1: Selection of Interaction Features Using F-Score -- 6.4.4.3. Step2: Selection of Features Using F-Score -- 6.4.4.4. Step3: Selection of Best Subset of Features Using Random Forest -- 6.4.5. Data Sampling for Training and Test -- 6.4.6. Class Balancing -- 6.4.7. Classifier (SVM) -- 6.5. Prediction Result -- 6.6. Model Evaluation -- 6.7. Conclusion -- References -- Chapter 7: Caravan Insurance Customer Profile Modeling with R -- 7.1. Introduction -- 7.2. Data Description and Initial Exploratory Data Analysis -- 7.2.1. Variable Correlations and Logistic Regression Analysis -- 7.3. Classifier Models of Caravan Insurance Holders -- 7.3.1. Overview of Model Building and Validating -- 7.3.2. Review of Four Classifier Methods -- 7.3.3. RP Model -- 7.3.4. Bagging Ensemble -- 7.3.5. Support Vector Machine -- 7.3.6. LR Classification -- 7.3.7. Comparison of Four Classifier Models: ROC and AUC -- 7.3.8. Model Comparison: Recall-Precision, Accuracy-v-Cut-off, and Computation Times -- 7.4. Discussion of Results and Conclusion.

Appendix A. Details of the Full Data Set Variables -- Appendix B. Customer Profile Data-Frequency of Binary Values -- Appendix C. Proportion of Caravan Insurance Holders vis-à-vis other Customer Profile Variables -- Appendix D. LR Model Details -- Appendix E. R Commands for Computation of ROC Curves for Each Model Using Validation Dataset -- Appendix F. Commands for Cross-Validation Analysis of Classifier Models -- References -- Chapter 8: Selecting Best Features for Predicting Bank Loan Default -- 8.1. Introduction -- 8.2. Business Problem -- 8.3. Data Extraction -- 8.4. Data Exploration and Preparation -- 8.4.1. Null Value Detection -- 8.4.2. Outlier Detection -- 8.5. Missing Imputation -- 8.5.1. Relevance Analysis -- 8.5.2. Data Set Balancing -- 8.5.3. Feature Selection -- 8.6. Modeling -- 8.7. Model Evaluation -- 8.8. Finding and Model Deployment -- 8.9. Lessons and Discussions -- Appendix. Selecting Best Features for Predicting Bank Loan Default -- References -- Chapter 9: A Choquet Integral Toolbox and Its Application in Customer Preference Analysis -- 9.1. Introduction -- 9.2. Background -- 9.2.1. Aggregation Functions -- 9.2.2. Choquet Integral -- 9.2.3. Fuzzy Measure Representation -- 9.2.4. Shapley Value and Interaction Index -- 9.3. Rfmtool Package -- 9.3.1. Installation -- 9.3.2. Toolbox Description -- 9.3.3. Preference Analysis Example -- 9.4. Case Study -- 9.4.1. Traveler Preference Study and Hotel Management -- 9.4.2. Data Collection and Experiment Design -- 9.4.3. Model Evaluation -- 9.4.4. Result Analysis -- 9.4.4.1. Preference Profile Construction -- 9.4.4.2. Interaction Behavior Analysis -- 9.4.5. Discussion -- 9.5. Conclusions -- References -- Chapter 10: A Real-Time Property Value Index Based on Web Data -- 10.1. Introduction -- 10.2. Housing Prices and Indices -- 10.3. A Data Mining Approach -- 10.3.1. Data Capture.

10.3.2. Geocoding -- 10.3.3. Price Evolution -- 10.4. Real Estate Pricing Models -- 10.4.1. Model 1: Hedonic Model Plus Smooth Term -- 10.4.2. Model 2: GWR Plus a Smooth Term -- 10.4.3. Relationship to Other Work -- 10.5. Conclusion -- Acknowledgments -- References -- Chapter 11: Predicting Seabed Hardness Using Random Forest in R -- 11.1. Introduction -- 11.2. Study Region and Data Processing -- 11.2.1. Study Region -- 11.2.2. Data Processing of Seabed Hardness -- 11.2.3. Predictors -- 11.3. Dataset Manipulation and Exploratory Analyses -- 11.3.1. Features of the Dataset -- 11.3.2. Exploratory Data Analyses -- 11.4. Application of RF for Predicting Seabed Hardness -- 11.5. Model Validation Using rfcv -- 11.6. Optimal Predictive Model -- 11.7. Application of the Optimal Predictive Model -- 11.8. Discussion and Conclusions -- 11.8.1. Selection of Relevant Predictors and the Consequences of Missing the Most Important Predictors -- 11.8.2. Issues with Searching for the Most Accurate Predictive Model Using RF -- 11.8.3. Predictive Accuracy of RF and Prediction Maps of Seabed Hardness -- 11.8.4. Limitations -- Appendix AA. Dataset of Seabed Hardness and 15 Predictors -- Appendix BA. R Function, rf.cv, Shows the Cross-Validated Prediction Performance of a Predictive Model -- References -- Chapter 12: Supervised Classification of Images, Applied to Plankton Samples Using R and Zooimage -- 12.1. Background -- 12.2. Challenges -- 12.3. Data Extraction and Exploration -- 12.4. Data Preprocessing -- 12.5. Modeling -- 12.6. Model Evaluation -- 12.7. Model Deployment -- 12.8. Lessons, Discussion, and Conclusions -- Acknowledgments -- References -- Chapter 13: Crime Analyses Using R -- 13.1. Introduction -- 13.2. Problem Definition -- 13.3. Data Extraction -- 13.4. Data Exploration and Preprocessing -- 13.5. Visualizations -- 13.6. Modeling.

13.7. Model Evaluation.