Feature extraction of weighted data for implicit variable selection

Luis Sánchez, Fernando Martínez, Germán Castellanos, Augusto Salazar

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Approaches based on obtaining relevant information from overwhelmingly large sets of measures have been recently adopted as an alternative to specialized features. In this work, we address the problem of finding a relevant subset of features and a suitable rotation (combined feature selection and feature extraction) as a weighted rotation. We focus our attention on two types of rotations: Weighted Principal Component Analysis and Weighted Regularized Discriminant Analysis. The objective function is the maximization of the J4 ratio. Tests were carried out on artificially generated classes, with several non-relevant features. Real data tests were also performed on segmentation of naildfold capillaroscopic images, and NIST-38 database (prototype selection).

Original languageEnglish
Title of host publicationComputer Analysis of Images and Patterns - 12th International Conference, CAIP 2007, Proceedings
Pages840-847
Number of pages8
DOIs
StatePublished - 2007
Event12th International Conference on Computer Analysis of Images and Patterns, CAIP 2007 - Vienna, Austria
Duration: Aug 27 2007Aug 29 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4673 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference12th International Conference on Computer Analysis of Images and Patterns, CAIP 2007
Country/TerritoryAustria
CityVienna
Period8/27/078/29/07

Keywords

  • Feature selection
  • LDA
  • PCA
  • Relevance

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Feature extraction of weighted data for implicit variable selection'. Together they form a unique fingerprint.

Cite this