Book Image

Predictive Analytics Using Rattle and Qlik Sense

By : Ferran Garcia Pagans, Fernando G Pagans
Book Image

Predictive Analytics Using Rattle and Qlik Sense

By: Ferran Garcia Pagans, Fernando G Pagans

Overview of this book

Table of Contents (16 chapters)
Predictive Analytics Using Rattle and Qlik Sense
Credits
About the Author
About the Reviewers
www.PacktPub.com
Preface
Index

Ensemble classifiers


Thomas G Dietterich defines Ensemble methods as follows:

"Ensemble methods are learning algorithms that construct a set of classifiers and then classify new data points by taking a (weighted) vote of their prediction."

You can get more information from http://web.engr.oregonstate.edu/~tgd/publications/mcs-ensembles.pdf.

Ensemble methods create a set of weak classifiers and combine them into a strong classifier. A weak classifier is a classifier that performs slightly better than a classifier that randomly guesses the prediction. Rattle offers two types of ensemble models: Random Forest and Boosting.

Boosting

Boosting is an ensemble method, so it creates a set of different classifiers. Imagine that you have m classifiers, we can define a classifier x as:

When we need to evaluate a new observation, we can calculate the average of all m tree's predictions using the following formula:

We can improve this evaluation by adding a weight to each tree, as shown here in this formula...