Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Education and Culture Executive Agency (EACEA). Neither the European Union nor EACEA can be held responsible for them.


Classification And Regression Trees

Users: 1 - Average Rating: 4.00

This is one of the pioneering books on classification and regression trees (CART), which constitutes an alternative to the construction of a prediction equation as it is done in linear regression. In CART, data are partitioned along the predictor variables into subsets with homogeneous values of the dependent variable, and then a decision tree emerges to make predictions from new observations in a similar vein as in the Guess Who? board game. But depending on the problem, the basic purpose can be either to produce an accurate classifier or to uncover the predictive structure of the problem, using Bayes rule as an underlying main tool. Along with two introductory and two intermediate illustrative chapters (with medical, ship and mass spectra classification examples) and how to get the right size of the tree and accurate estimates of misclassification probability, some issues as splitting rules (variable misclassification costs by generalizing Gini criterion or by prior alteration), best splitting and Boolean combinations are dealt with to improve flexibility, power, and efficient of classification trees. Then authors proceed to adapt their technique to (claimed simpler) regression trees, so the final tree procedures (Bayesian partitions and optimal pruning using joint distributions, and topics related with resampling the learning sample and its size tending to infinity) can be developed for the general framework.


Machine Learning

Target Group:


Cite as:
L. Breiman, J. Friedman, R. Olshen, C. Stone (1984), Classification and Regression Trees. Routledge (Taylor and Francis imprint), New York. ISBN 9781315139470 (ebook).

Author of the review:
Pablo Guerrero-Garcia
University of Malaga


You have to login to leave a comment. If you are not registered click here

Davide Ilardi

Still a good introduction to classification and regression trees with a variety of examples! Suggested for people with some statistics and probability backgrounds.