Abstract of FKI-221-97

Document-Name:  fki-221-97.ps.gz
Title:	        Feature Selection by Means of a Feature Weighting Approach
Authors:	Matthias Scherf & Wilfried Brauer
Revision-Date:	1997/4/1
Category:	Technical Report (Forschungsberichte Künstliche Intelligenz)
Abstract:	Selecting a set of features which is optimal for a given
                classification task is one of the central problems in machine 
                learning. We address the problem using the flexible and 
                robust filter technique EUBAFES. EUBAFES is based on a feature
                weighting approach which computes binary feature weights and 
                therefore a solution in the feature selection sense and
                also gives detailed information about feature relevance by
                continuous weights. Moreover the user gets not only one but 
                several potentially optimal feature subsets which is important
                for filter-based feature selection algorithms since it gives 
                the flexibility to use even complex classifiers by the 
                application of a combined filter/wrapper approach. We applied 
                EUBAFES on a number of artificial and real world data sets and
                used radial basis function networks to examine the impact of 
                the feature subsets to classifier accuracy and complexity.
Keywords:       Feature Weighting, Feature Selection, RBF-Classifier
Size:		22 pages
Language:       English
ISSN:		0941-6358
Copyright:	The ``Forschungsberichte Künstliche Intelligenz''
		series includes primarily preliminary publications,
		specialized partial results, and supplementary
		material. In the interest of a subsequent final
		publication these reports should not be copied. All
		rights and the responsability for the contents of the
		report are with the authors, which would appreciate
		critical comments.

Gerhard Weiss