Ensemble Methods

• Published in: Machine Learning in Python pp. 211 - 253
• Main Author: Bowles, Michael
• Format: Book Chapter
• Language: English
• Published: United States John Wiley & Sons, Incorporated 2015; John Wiley & Sons, Inc
• Subjects: bagging algorithm; base learners; binary decision trees; bootstrap aggregation; ensemble methods; gradient boosting algorithm; multivariable tree; Python code; random forests
• ISBN: 1118961749; 9781118961742
• DOI: 10.1002/9781119183600.ch6

Ensemble methods stem from the observation that multiple models give better performance than a single model if the models are somewhat independent of one another. The key with ensemble methods is to develop an algorithm approach to generate numerous somewhat independent models that will then be combined into an ensemble. This chapter shows the reader how the most popular methods accomplish this. It teaches the mechanics of the most popular ensemble methods, and outlines the basic structure of the algorithms and demonstrates the algorithms in Python code to give a firm understanding of their workings. The chapter primarily uses binary decision trees as base learners. Some widely used upper‐level algorithms are known as bagging, gradient boosting, and random forests are coded in this chapter to provide a better understand the options, input variables, nominal starting values, and so on for the Python packages for these algorithms.