Bayesian Feature Selection for Clustering Problems

Eduardo R. Hruschka (), Estevam R. Hruschka (), Thiago F. Covões () and Nelson F. F. Ebecken ()
Additional contact information
Eduardo R. Hruschka: Catholic University of Santos (UniSantos), Brazil
Estevam R. Hruschka: Federal University of São Carlos, Brazil
Thiago F. Covões: Catholic University of Santos (UniSantos), Brazil
Nelson F. F. Ebecken: COPPE / Federal University of Rio de Janeiro, Brazil

Journal of Information & Knowledge Management (JIKM), 2006, vol. 05, issue 04, 315-327

Abstract: Bayesian methods have been successfully used for feature selection in many supervised learning tasks. In this paper, the adaptation of such methods for unsupervised learning (clustering) is investigated. We adopt an algorithm that iterates between clustering (assuming that the number of clusters is unknowna priori) and feature selection. From this standpoint, two Bayesian approaches for feature selection are addressed: (i) Naïve Bayes Wrapper (NBW), and (ii) Markov Blanket Filter (MBF) obtained from the construction of Bayesian networks. Experiments in ten datasets illustrate the performance of each proposed method. Advantages of feature selection are demonstrated by comparing the results obtained from Bayesian feature selection with the results achieved without any kind of feature selection, i.e., using all the available features. In most of the performed experiments, NBW and MBF have allowed reducing the number of features, while providing good quality partitions in relation to those found by means of the full set of features. Also, NBW has outperformed its Bayesian feature selection counterpart (MBF) in most of the assessed datasets, mainly when the cardinality of the selected feature subset is taken into consideration.

Keywords: Feature selection; clustering; Naïve Bayes; Bayesian networks (search for similar items in EconPapers)
Date: 2006
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.worldscientific.com/doi/abs/10.1142/S0219649206001578
Access to full text is restricted to subscribers

Related works:
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:wsi:jikmxx:v:05:y:2006:i:04:n:s0219649206001578

Ordering information: This journal article can be ordered from

DOI: 10.1142/S0219649206001578

Access Statistics for this article

Journal of Information & Knowledge Management (JIKM) is currently edited by Professor Suliman Hawamdeh

More articles in Journal of Information & Knowledge Management (JIKM) from World Scientific Publishing Co. Pte. Ltd.
Bibliographic data for series maintained by Tai Tone Lim ().