Affordable Access

Access to the full text

Assessing variable importance in clustering: a new method based on unsupervised binary decision trees

Authors
  • Badih, Ghattas1
  • Pierre, Michel1, 2
  • Laurent, Boyer2
  • 1 Aix Marseille Université, CNRS, Centrale Marseille, I2M UMR 7373, Marseille, 13453, France , Marseille (France)
  • 2 Aix Marseille Université, SPMC EA3279, Marseille, 13385, France , Marseille (France)
Type
Published Article
Journal
Computational Statistics
Publisher
Springer Berlin Heidelberg
Publication Date
Jan 05, 2019
Volume
34
Issue
1
Pages
301–321
Identifiers
DOI: 10.1007/s00180-018-0857-0
Source
Springer Nature
Keywords
License
Yellow

Abstract

We consider different approaches for assessing variable importance in clustering. We focus on clustering using binary decision trees (CUBT), which is a non-parametric top-down hierarchical clustering method designed for both continuous and nominal data. We suggest a measure of variable importance for this method similar to the one used in Breiman’s classification and regression trees. This score is useful to rank the variables in a dataset, to determine which variables are the most important or to detect the irrelevant ones. We analyze both stability and efficiency of this score on different data simulation models in the presence of noise, and compare it to other classical variable importance measures. Our experiments show that variable importance based on CUBT is much more efficient than other approaches in a large variety of situations.

Report this publication

Statistics

Seen <100 times