The Support Feature Machine for Classifying with the Least Number of Features

Sascha Klement, Thomas Martinetz

Abstract

We propose the so-called Support Feature Machine (SFM) as a novel approach to feature selection for classification, based on minimisation of the zero norm of a separating hyperplane. Thus, a classifier with inherent feature selection capabilities is obtained within a single training run. Results on toy examples demonstrate that this method is able to identify relevant features very effectively.
Original languageEnglish
Title of host publicationArtificial Neural Networks -- ICANN 2010
EditorsKonstantinos Diamantaras, Wlodek Duch, Lazaros S. Iliadis
Number of pages6
Volume6353
Place of PublicationBerlin, Heidelberg
PublisherSpringer Berlin Heidelberg
Publication date2010
Pages88-93
ISBN (Print)978-3-642-15821-6
ISBN (Electronic)978-3-642-15822-3
DOIs
Publication statusPublished - 2010
Event20th International Conference Artificial Neural Networks
- Thessaloniki, Greece
Duration: 15.09.201018.09.2010

Fingerprint

Dive into the research topics of 'The Support Feature Machine for Classifying with the Least Number of Features'. Together they form a unique fingerprint.

Cite this