A similarity measure to assess the stability of classification trees

Authors:
Bénédicte Briand;Gilles R. Ducharme;Vanessa Parache;Catherine Mercat-Rommens
Affiliations:
IRSN, Laboratoire d'Etudes Radioécologiques en milieux Continental et Marin, DEI/SESURE/LERCM, Cadarache, bít. 153, BP 3, 13115 Saint-Paul-lez-Durance, France;Equipe de probabilités et statistique, I3M, Université Montpellier II, cc 051, Place Eugène Bataillon, 34095 Montpellier Cedex 5, France;IRSN, Laboratoire d'Etudes Radioécologiques en milieux Continental et Marin, DEI/SESURE/LERCM, Cadarache, bít. 153, BP 3, 13115 Saint-Paul-lez-Durance, France;IRSN, Laboratoire d'Etudes Radioécologiques en milieux Continental et Marin, DEI/SESURE/LERCM, Cadarache, bít. 153, BP 3, 13115 Saint-Paul-lez-Durance, France
Venue:
Computational Statistics & Data Analysis
Year:
2009

Citing 7
Cited 1

Simplifying decision trees

International Journal of Man-Machine Studies - Special Issue: Knowledge Acquisition for Knowledge-based Systems. Part 5
Bagging predictors

Machine Learning
A Comparative Analysis of Methods for Pruning Decision Trees

IEEE Transactions on Pattern Analysis and Machine Intelligence
Random Forests

Machine Learning
An Empirical Comparison of Pruning Methods for Decision Tree Induction

Machine Learning
Empirical characterization of random forest variable importance measures

Computational Statistics & Data Analysis
CART algorithm for spatial data: Application to environmental and ecological data

Computational Statistics & Data Analysis

Binary trees for dissimilarity data

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.03

Visualization

Abstract

It has been recognized that Classification trees (CART) are unstable; a small perturbation in the input variables or a fresh sample can lead to a very different classification tree. Some approaches exist that try to correct this instability. However, their benefits can, at present, be appreciated only qualitatively. A similarity measure between two classification trees is introduced that can measure their closeness. Its usefulness is illustrated with synthetic data on the impact of radioactivity deposit through the environment. In this context, a modified node level stabilizing technique, referred to as the NLS-REP method, is introduced and shown to be more stable than the classical CART method.