Introduction to Statistical and Machine Learning Methods for Data Science

Table of contents

1. About This Book
2. About These Authors
3. Acknowledgments
4. Foreword
5. Chapter 1: Introduction to Data Science
   1. Chapter Overview
   2. Data Science
      1. Mathematics and Statistics
      2. Computer Science
      3. Domain Knowledge
      4. Communication and Visualization
      5. Hard and Soft Skills
   3. Data Science Applications
   4. Data Science Lifecycle and the Maturity Framework
      1. Understand the Question
      2. Collect the Data
      3. Explore the Data
      4. Model the Data
      5. Provide an Answer
   5. Advanced Analytics in Data Science
   6. Data Science Practical Examples
      1. Customer Experience
      2. Revenue Optimization
      3. Network Analytics
      4. Data Monetization
   7. Summary
   8. Additional Reading
6. Chapter 2: Data Exploration and Preparation
   1. Chapter Overview
   2. Introduction to Data Exploration
      1. Nonlinearity
      2. High Cardinality
      3. Unstructured Data
      4. Sparse Data
      5. Outliers
      6. Mis-scaled Input Variables
   3. Introduction to Data Preparation
      1. Representative Sampling
      2. Event-based Sampling
      3. Partitioning
      4. Imputation
      5. Replacement
      6. Transformation
      7. Feature Extraction
      8. Feature Selection
   4. Model Selection
      1. Model Generalization
      2. Bias–Variance Tradeoff
   5. Summary
7. Chapter 3: Supervised Models – Statistical Approach
   1. Chapter Overview
   2. Classification and Estimation
   3. Linear Regression
      1. Use Case: Customer Value
   4. Logistic Regression
      1. Use Case: Collecting Predictive Model
   5. Decision Tree
      1. Use Case: Subscription Fraud
   6. Summary
8. Chapter 4: Supervised Models – Machine Learning Approach
   1. Chapter Overview
   2. Supervised Machine Learning Models
   3. Ensemble of Trees
      1. Random Forest
      2. Gradient Boosting
      3. Use Case: Usage Fraud
   4. Neural Network
      1. Use Case: Bad Debt
   5. Summary
9. Chapter 5: Advanced Topics in Supervised Models
   1. Chapter Overview
   2. Advanced Machine Learning Models and Methods
   3. Support Vector Machines
      1. Use Case: Fraud in Prepaid Subscribers
   4. Factorization Machines
      1. Use Case: Recommender Systems Based on Customer Ratings in Retail
   5. Ensemble Models
      1. Use Case Study: Churn Model for Telecommunications
   6. Two-stage Models
      1. Use Case: Anti-attrition
   7. Summary
   8. Additional Reading
10. Chapter 6: Unsupervised Models—Structured Data
    1. Chapter Overview
    2. Clustering
    3. Hierarchical Clustering
       1. Use Case: Product Segmentation
    4. Centroid-based Clustering (k-means Clustering)
       1. Use Case: Customer Segmentation
    5. Self-organizing Maps
       1. Use Case Study: Insolvent Behavior
    6. Cluster Evaluation
       1. Cluster Profiling
       2. Additional Topics
    7. Summary
    8. Additional Reading
11. Chapter 7: Unsupervised Models—Semi Structured Data
    1. Chapter Overview
    2. Association Rules Analysis
       1. Market Basket Analysis
       2. Confidence and Support Measures
       3. Use Case: Product Bundle Example
       4. Expected Confidence and Lift Measures
       5. Association Rules Analysis Evaluation
       6. Use Case: Product Acquisition
    3. Sequence Analysis
       1. Use Case: Next Best Offer
    4. Link Analysis
       1. Use Case: Product Relationships
    5. Path Analysis
       1. Use Case Study: Online Experience
    6. Text Analytics
       1. Use Case Study: Call Center Categorization
    7. Summary
    8. Additional Reading
12. Chapter 8: Advanced Topics in Unsupervised Models
    1. Chapter Overview
    2. Network Analysis
       1. Network Subgraphs
       2. Network Metrics
       3. Use Case: Social Network Analysis to Reduce Churn in Telecommunications
    3. Network Optimization
       1. Network Algorithms
       2. Use Case: Smart Cities – Improving Commuting Routes
    4. Summary
13. Chapter 9: Model Assessment and Model Deployment
    1. Chapter Overview
    2. Methods to Evaluate Model Performance
       1. Speed of Training
       2. Speed of Scoring
       3. Business Knowledge
       4. Fit Statistics
       5. Data Splitting
       6. K-fold Cross-validation
       7. Goodness-of-fit Statistics
       8. Confusion Matrix
       9. ROC Curve
       10. Model Evaluation
    3. Model Deployment
       1. Challenger Models
       2. Monitoring
    4. Model Operationalization
    5. Summary