Local Feature Selection for Data Classification

Typical feature selection methods choose an optimal global feature subset that is applied over all regions of the sample space. In contrast, in this paper we propose a novel localized feature selection (LFS) approach whereby each region of the sample space is associated with its own distinct optimiz...

Full description

Bibliographic Details

Published in:	IEEE transactions on pattern analysis and machine intelligence Vol. 38; no. 6; pp. 1217 - 1227
Main Authors:	Armanfard, Narges, Reilly, James P., Komeili, Majid
Format:	Journal Article
Language:	English
Published:	United States IEEE 01.06.2016 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:	Classification Feature extraction Frequency modulation Linear programming Local Feature Selection Manifolds Pareto optimization Training classification Local feature selection linear programming
ISSN:	0162-8828, 1939-3539, 2160-9292, 1939-3539
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!