Dimensionality reduction strategies for the design of human machine interface signal classifiers

Lalit Gupta, Srinivas Kota, Swetha Murali, Dennis Molfese, Ravi Vaidyanathan

Research output: Contribution to journalConference article

1 Citation (Scopus)

Abstract

The goal in this paper is to overcome the dimensionality problem related to designing human-machineinterface (HMI) signal classifiers. The dimension is decreased by selecting a small set of linear combination of the input space features using the principal components transform (PCT) and the discrete cosine transform (DCT). Issues dealing with the selection of the basis vectors of the PCT and DCT for multi-class classification problems are addressed and four different classdependant ranking criteria are introduced to select basis vectors from the transformed training vectors in the PCT and DCT domains. The application and evaluation of the resulting PCT and DCT based multivariate classification strategies are demonstrated by classifying ear-pressure signals and event related potentials. The signals in these experiments are typical of control signals used in HMI applications and are also typical of those in which the dimensionality problem occurs. Based on the evaluations and comparisons, it is concluded that the PCT and the DCT based strategies developed in this paper offer viable solutions to overcome the dimensionality problem that frequently plagues the design of practical HMI signal classifiers.

Original languageEnglish (US)
Article number4811659
Pages (from-to)2432-2436
Number of pages5
JournalConference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
DOIs
StatePublished - Dec 1 2008
Event2008 IEEE International Conference on Systems, Man and Cybernetics, SMC 2008 - Singapore, Singapore
Duration: Oct 12 2008Oct 15 2008

Fingerprint

Discrete cosine transforms
Classifiers
Experiments

Keywords

  • Dimensionality reduction
  • Discrete cosine transform
  • Ear-pressure signals
  • Event related potentials
  • HMI signal classification
  • Principal components transform

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Human-Computer Interaction

Cite this

Dimensionality reduction strategies for the design of human machine interface signal classifiers. / Gupta, Lalit; Kota, Srinivas; Murali, Swetha; Molfese, Dennis; Vaidyanathan, Ravi.

In: Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, 01.12.2008, p. 2432-2436.

Research output: Contribution to journalConference article

@article{fe46a7ea2f4b42d697343bf7fbf689b0,
title = "Dimensionality reduction strategies for the design of human machine interface signal classifiers",
abstract = "The goal in this paper is to overcome the dimensionality problem related to designing human-machineinterface (HMI) signal classifiers. The dimension is decreased by selecting a small set of linear combination of the input space features using the principal components transform (PCT) and the discrete cosine transform (DCT). Issues dealing with the selection of the basis vectors of the PCT and DCT for multi-class classification problems are addressed and four different classdependant ranking criteria are introduced to select basis vectors from the transformed training vectors in the PCT and DCT domains. The application and evaluation of the resulting PCT and DCT based multivariate classification strategies are demonstrated by classifying ear-pressure signals and event related potentials. The signals in these experiments are typical of control signals used in HMI applications and are also typical of those in which the dimensionality problem occurs. Based on the evaluations and comparisons, it is concluded that the PCT and the DCT based strategies developed in this paper offer viable solutions to overcome the dimensionality problem that frequently plagues the design of practical HMI signal classifiers.",
keywords = "Dimensionality reduction, Discrete cosine transform, Ear-pressure signals, Event related potentials, HMI signal classification, Principal components transform",
author = "Lalit Gupta and Srinivas Kota and Swetha Murali and Dennis Molfese and Ravi Vaidyanathan",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/ICSMC.2008.4811659",
language = "English (US)",
pages = "2432--2436",
journal = "Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics",
issn = "1062-922X",

}

TY - JOUR

T1 - Dimensionality reduction strategies for the design of human machine interface signal classifiers

AU - Gupta, Lalit

AU - Kota, Srinivas

AU - Murali, Swetha

AU - Molfese, Dennis

AU - Vaidyanathan, Ravi

PY - 2008/12/1

Y1 - 2008/12/1

N2 - The goal in this paper is to overcome the dimensionality problem related to designing human-machineinterface (HMI) signal classifiers. The dimension is decreased by selecting a small set of linear combination of the input space features using the principal components transform (PCT) and the discrete cosine transform (DCT). Issues dealing with the selection of the basis vectors of the PCT and DCT for multi-class classification problems are addressed and four different classdependant ranking criteria are introduced to select basis vectors from the transformed training vectors in the PCT and DCT domains. The application and evaluation of the resulting PCT and DCT based multivariate classification strategies are demonstrated by classifying ear-pressure signals and event related potentials. The signals in these experiments are typical of control signals used in HMI applications and are also typical of those in which the dimensionality problem occurs. Based on the evaluations and comparisons, it is concluded that the PCT and the DCT based strategies developed in this paper offer viable solutions to overcome the dimensionality problem that frequently plagues the design of practical HMI signal classifiers.

AB - The goal in this paper is to overcome the dimensionality problem related to designing human-machineinterface (HMI) signal classifiers. The dimension is decreased by selecting a small set of linear combination of the input space features using the principal components transform (PCT) and the discrete cosine transform (DCT). Issues dealing with the selection of the basis vectors of the PCT and DCT for multi-class classification problems are addressed and four different classdependant ranking criteria are introduced to select basis vectors from the transformed training vectors in the PCT and DCT domains. The application and evaluation of the resulting PCT and DCT based multivariate classification strategies are demonstrated by classifying ear-pressure signals and event related potentials. The signals in these experiments are typical of control signals used in HMI applications and are also typical of those in which the dimensionality problem occurs. Based on the evaluations and comparisons, it is concluded that the PCT and the DCT based strategies developed in this paper offer viable solutions to overcome the dimensionality problem that frequently plagues the design of practical HMI signal classifiers.

KW - Dimensionality reduction

KW - Discrete cosine transform

KW - Ear-pressure signals

KW - Event related potentials

KW - HMI signal classification

KW - Principal components transform

UR - http://www.scopus.com/inward/record.url?scp=69949141182&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69949141182&partnerID=8YFLogxK

U2 - 10.1109/ICSMC.2008.4811659

DO - 10.1109/ICSMC.2008.4811659

M3 - Conference article

AN - SCOPUS:69949141182

SP - 2432

EP - 2436

JO - Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics

JF - Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics

SN - 1062-922X

M1 - 4811659

ER -