Most data mining opportunities involve machine learning and often come with greater financial rewards. This book will help you bring the power of machine learning techniques into your data mining work. By the end of the book, you will be able to create accurate predictive models for data mining.

• Title Page
• Copyright and Credits
• Contributors
• About Packt
• Table of Contents
• Preface
• Introducing Machine Learning Predictive Models
  • Characteristics of machine learning predictive models
  • Types of machine learning predictive models
  • Working with neural networks
    • Advantages of neural networks
    • Disadvantages of neural networks
    • Representing the errors
    • Types of neural network models
    • Multi-layer perceptron
      • Why are weights important?
      • An example representation of a multilayer perceptron model
    • The linear regression model
  • A sample neural network model
    • Feed-forward backpropagation
    • Model training ethics
  • Summary
• Getting Started with Machine Learning
  • Demonstrating a neural network
    • Running a neural network model
    • Interpreting results
      • Analyzing the accuracy of the model
    • Model performance on testing partition
  • Support Vector Machines
    • Working with Support Vector Machines
      • Kernel transformation
        • But what is the best solution?
        • Types of kernel functions
  • Demonstrating SVMs
    • Interpreting the results
    • Trying additional solutions
  • Summary
• Understanding Models
  • Models
    • Statistical models
    • Decision tree models
    • Machine learning models
  • Using graphs to interpret machine learning models
  • Using statistics to interpret machine learning models
    • Understanding the relationship between a continuous predictor and a categorical outcome variable
  • Using decision trees to interpret machine learning models
  • Summary
• Improving Individual Models
  • Modifying model options
  • Using a different model to improve results
  • Removing noise to improve models
    • How to remove noise
  • Doing additional data preparation
    • Preparing the data
  • Balancing data
    • The need for balancing data
    • Implementing balance in data
  • Summary 
• Advanced Ways of Improving Models
  • Combining models
    • Combining by voting
    • Combining by highest confidence
    • Implementing combining models
      • Combining models in Modeler
      • Combining models outside Modeler
  • Using propensity scores
    • Implementations of propensity scores
  • Meta-level modeling
  • Error modeling
  • Boosting and bagging
    • Boosting
    • Bagging
  • Predicting continuous outcomes
  • Summary
• Other Books You May Enjoy
• Index