Predictive Analytics for the Modern Enterprise

Table of contents

1. Preface
   1. Who Is This Book For?
   2. How This Book Is Organized
   3. Conventions Used in This Book
   4. Using Code Examples
   5. O’Reilly Online Learning
   6. How to Contact Us
   7. Acknowledgments
2. 1. Data Analytics in the Modern Enterprise
   1. The Evolution of Data Analytics
   2. Different Types of Data Analytics
      1. Descriptive Analytics
      2. Diagnostic Analytics
      3. Predictive Analytics
      4. Prescriptive Analytics
   3. Knowledge Acquisition, Machine Learning, and the Role of Predictive Analytics
   4. Tools, Frameworks, and Platforms in the Predictive Analytics World
      1. Languages and Libraries
      2. Services
   5. Conclusion
3. 2. Predictive Analytics: An Operational Necessity
   1. The Move from “Data Producing” to “Data Driven”
   2. Challenges to Using Predictive Analytics
      1. People
      2. Data
      3. Technology
   3. Vertical Industry Use Cases for Predictive Analytics
      1. Finance
      2. Healthcare
      3. Automotive
      4. Entertainment
   4. Conclusion
4. 3. The Mathematics and Algorithms Behind Predictive Analytics
   1. Statistics and Linear Algebra
   2. Regression
      1. What Is Regression Analysis?
      2. Regression Techniques
      3. R-squared and P-value
      4. Selecting a Regression Model
   3. Decision Trees
      1. Training Decision Trees
      2. Using Decision Trees to Solve Regression Problems: Regression Trees
      3. Tuning Decision Trees
   4. Other Algorithms
      1. Random Forests
      2. Neural Networks
      3. Support Vector Machines
      4. Naive Bayes Classifier
   5. Other Learning Patterns in Machine Learning
   6. Conclusion
5. 4. Working with Data
   1. Understanding Data
   2. Data Preprocessing and Feature Engineering
      1. Handling Missing Data
      2. Categorical Data Encoding
      3. Data Transformation
      4. Outlier Management
      5. Handling Imbalanced Data
      6. Combining Data
      7. Feature Selection
      8. Splitting Preprocessed Data
   3. Understanding Bias
   4. The Predictive Analytics Pipeline
      1. The Data Stage
      2. The Model Stage
      3. The Serving Stage
      4. Other Components
   5. Selecting the Right Model
   6. Conclusion
6. 5. Python and scikit-learn for Predictive Analytics
   1. Anaconda and Jupyter Notebooks
   2. NumPy in Python
      1. Introduction to NumPy
      2. Generating Arrays
      3. Array Slicing
      4. Array Transformation
      5. Other Array Operations
      6. Exploring a Business Example Using Pandas
   3. Pandas in Python
      1. Import and View Data
      2. Visualize the Data
      3. Data Cleaning and Modification
      4. Reading from Different Data Sources
      5. Data Filtering and Grouping
   4. Scikit-learn
      1. Training and Predicting with a Linear Regression Model
      2. Using a Random Forest Classifier
      3. Training a Decision Tree
      4. A Clustering Example (Unsupervised Learning)
   5. Conclusion
7. 6. TensorFlow and Keras for Predictive Analytics
   1. TensorFlow Fundamentals
   2. Linear Regression Using TensorFlow
      1. Data Preparation
      2. Model Creation and Training
      3. Predictions and Model Evaluation
   3. Deep Neural Networks in TensorFlow
   4. Conclusion
8. 7. Predictive Analytics for Business Problem-Solving
   1. Prediction-Based Optimal Retail Price Recommendations
      1. Using a Simple Linear Regression Model
      2. Using a Polynomial Regression Model
      3. Using Multivariate Regression
   2. An Introduction to Recommender Systems
      1. Building Recommender Systems Using surprise scikit in Python
   3. Credit Card Fraud Classification
      1. Credit Card Fraud Baseline Analysis Using Artificial Neural Networks
      2. Credit Card Fraud Weighted Analysis Using Artificial Neural Networks
      3. Credit Card Analysis with Multiple Hidden Layers in the Artificial Neural Network
   4. Conclusion
9. 8. Exploring AWS Cloud Provider Services for AI/ML
   1. To Cloud or Not to Cloud
   2. Exploring AWS SageMaker
      1. Prerequisites
      2. Data Ingest and Exploration
      3. Data Transformation
      4. Model Training and Prediction
      5. Cleanup
   3. Exploring Amazon Forecast
      1. Import Data
      2. Train the Predictor
      3. Create a Forecast
      4. What-if Analysis
      5. Cleanup
   4. Conclusion
10. 9. Food for Thought
    1. A Few More Use Cases
       1. Navigation and Traffic Management
       2. Credit Scoring
    2. The Social Impact of Predictions
    3. Conclusion
11. About the Author