Another variant of robust fuzzy PCA with initial membership estimation

Gyeongyong Heo, Seong Hoon Kim, Young Woon Woo, Imgeun Lee

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

Abstract

Principal component analysis (PCA) is a well-known method for dimensionality reduction and feature extraction. PCA has been applied in many areas successfully, however, one of its problems is noise sensitivity due to the use of sum-square-error. Several variants of PCA have been proposed to resolve the problem and, among the variants, improved robust fuzzy PCA (RF-PCA2) demonstrated promising results. RF-PCA2, however, still can be affected by noise due to equal initial membership values for all data points. The fact that RF-PCA2 is still based on sum-square-error is another reason for noise sensitivity. In this paper, a variant of RF-PCA2 called RF-PCA3 is proposed. The proposed algorithm modifies the objective function of RF-PCA2 to allow some increase of sum-square-error and calculates initial membership values using data distribution. RF-PCA3 outperforms RF-PCA2, which is supported by experimental results.

Original languageEnglish
Title of host publicationIntelligent Information and Database Systems - Third International Conference, ACIIDS 2011, Proceedings
Pages120-129
Number of pages10
EditionPART 2
DOIs
StatePublished - 2011
Event3rd International Conference on Intelligent Information and Database Systems, ACIIDS 2011 - Daegu, Korea, Republic of
Duration: 20 Apr 201122 Apr 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume6592 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference3rd International Conference on Intelligent Information and Database Systems, ACIIDS 2011
Country/TerritoryKorea, Republic of
CityDaegu
Period20/04/1122/04/11

Keywords

  • KD-tree
  • Membership initialization
  • Nearest neighbor
  • Noise sensitivity
  • Principal component analysis

Fingerprint

Dive into the research topics of 'Another variant of robust fuzzy PCA with initial membership estimation'. Together they form a unique fingerprint.

Cite this