Number of the records: 1
A New Measure of Feature Selection Algorithms’ Stability
- 1.
SYSNO ASEP 0337780 Document Type C - Proceedings Paper (int. conf.) R&D Document Type Conference Paper Title A New Measure of Feature Selection Algorithms’ Stability Title Nová míra stability algoritmů výběru příznaků Author(s) Novovičová, Jana (UTIA-B)
Somol, Petr (UTIA-B) RID
Pudil, Pavel (UTIA-B) RIDSource Title ICDMW '09: Proceedings of the 2009 IEEE International Conference on Data Mining Workshops, OEDM '09: Workshop on Optimization Based Methods for Emerging Data Mining Problems. - Washington, DC, USA : IEEE Computer Society, 2009 - ISBN 978-0-7695-3902-7 Pages s. 382-387 Number of pages 6 s. Publication form - - - Action ICDMW '09: International Conference on Data Mining Workshops Event date 06.12.2009-09.12.2009 VEvent location Miami, Florida Country US - United States Event type WRD Language eng - English Country US - United States Keywords feature selection ; stability ; measure Subject RIV BD - Theory of Information R&D Projects 1M0572 GA MŠMT - Ministry of Education, Youth and Sports (MEYS) GA102/07/1594 GA ČR - Czech Science Foundation (CSF) GA102/08/0593 GA ČR - Czech Science Foundation (CSF) CEZ AV0Z10750506 - UTIA-B (2005-2011) Annotation Stability or robustness of feature selection methods is a topic of recent interest. A new stability measure based on the Shannon entropy is proposed in this paper to evaluate the overall occurrence of individual features in selected subsets of possibly varying cardinality. We compare the new measure to stability measures proposed recently by Somol et al. The new measure is computationally very efficient and adds another type of insight into the stability problem. All considered measures have been used to compare the stability of several feature selection methods (individually best ranking, sequential forward selection, sequential forward floating selection and dynamic oscillating search) on a set of examples. Workplace Institute of Information Theory and Automation Contact Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201. Year of Publishing 2010
Number of the records: 1