Machine Learning with scikit-learn Quick Start Guide

The AdaBoost classifier

In the section, you will learn how the AdaBoost classifier works internally, and how the concept of boosting might be used to give you better results. Boosting is a form of ensemble machine learning, in which a machine learning model learns from the mistakes of the models that were previously built, thereby increasing its final prediction accuracy.

AdaBoost stands for Adaptive Boosting, and is a boosting algorithm in which a lot of importance is given to the rows of data that the initial predictive model got wrong. This ensures that the next predictive model will not make the same mistakes.

The process by which the AdaBoost algorithm works is illustrated in the following diagram:

An outline of the AdaBoost algorithm ...