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

Badih Ghattas, Pierre Michel and Boyer Laurent
Additional contact information
Boyer Laurent: CEReSS - Centre d'études et de recherche sur les services de santé et la qualité de vie - AMU - Aix Marseille Université

Post-Print from HAL

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.

Keywords: Variables ranking; Unsupervised learning; CUBT; Deviance; Variable importance (search for similar items in EconPapers)
Date: 2019-03
New Economics Papers: this item is included in nep-cmp and nep-ecm
Note: View the original document on HAL open archive server: https://amu.hal.science/hal-02007388
References: Add references at CitEc
Citations:

Published in Computational Statistics, 2019, 34 (1), pp.301-321. ⟨10.1007/s00180-018-0857-0⟩

Downloads: (external link)
https://amu.hal.science/hal-02007388/document (application/pdf)

Related works:
Journal Article: Assessing variable importance in clustering: a new method based on unsupervised binary decision trees (2019)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:journl:hal-02007388

DOI: 10.1007/s00180-018-0857-0

Access Statistics for this paper

More papers in Post-Print from HAL
Bibliographic data for series maintained by CCSD ().