Pairwise diversity ranking of polychotomous features for ensemble physiological signal classifiers

Lalit Gupta, Srinivas Kota, Dennis L. Molfese, Ravi Vaidyanathan

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

It is well known that fusion classifiers for physiological signal classification with diverse components (classifiers or data sets) outperform those with less diverse components. Determining component diversity, therefore, is of the utmost importance in the design of fusion classifiers that are often employed in clinical diagnostic and numerous other pattern recognition problems. In this article, a new pairwise diversity-based ranking strategy is introduced to select a subset of ensemble components, which when combined will be more diverse than any other component subset of the same size. The strategy is unified in the sense that the components can be classifiers or data sets. Moreover, the classifiers and data sets can be polychotomous. Classifier-fusion and data-fusion systems are formulated based on the diversity-based selection strategy, and the application of the two fusion strategies are demonstrated through the classification of multichannel event-related potentials. It is observed that for both classifier and data fusion, the classification accuracy tends to increase/decrease when the diversity of the component ensemble increases/decreases. For the four sets of 14-channel event-related potentials considered, it is shown that data fusion outperforms classifier fusion. Furthermore, it is demonstrated that the combination of data components that yield the best performance, in a relative sense, can be determined through the diversity-based selection strategy.

Original languageEnglish (US)
Pages (from-to)655-662
Number of pages8
JournalProceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine
Volume227
Issue number6
DOIs
StatePublished - Jun 1 2013

Fingerprint

Classifiers
Evoked Potentials
Data fusion
Information Systems
Datasets
Pattern recognition

Keywords

  • Classifier fusion
  • Data fusion
  • Diversity
  • Ensemble classifiers
  • Event-related potentials

ASJC Scopus subject areas

  • Mechanical Engineering

Cite this

Pairwise diversity ranking of polychotomous features for ensemble physiological signal classifiers. / Gupta, Lalit; Kota, Srinivas; Molfese, Dennis L.; Vaidyanathan, Ravi.

In: Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine, Vol. 227, No. 6, 01.06.2013, p. 655-662.

Research output: Contribution to journalArticle

@article{4f74d10c4c6b41cb8d0a41d301229148,
title = "Pairwise diversity ranking of polychotomous features for ensemble physiological signal classifiers",
abstract = "It is well known that fusion classifiers for physiological signal classification with diverse components (classifiers or data sets) outperform those with less diverse components. Determining component diversity, therefore, is of the utmost importance in the design of fusion classifiers that are often employed in clinical diagnostic and numerous other pattern recognition problems. In this article, a new pairwise diversity-based ranking strategy is introduced to select a subset of ensemble components, which when combined will be more diverse than any other component subset of the same size. The strategy is unified in the sense that the components can be classifiers or data sets. Moreover, the classifiers and data sets can be polychotomous. Classifier-fusion and data-fusion systems are formulated based on the diversity-based selection strategy, and the application of the two fusion strategies are demonstrated through the classification of multichannel event-related potentials. It is observed that for both classifier and data fusion, the classification accuracy tends to increase/decrease when the diversity of the component ensemble increases/decreases. For the four sets of 14-channel event-related potentials considered, it is shown that data fusion outperforms classifier fusion. Furthermore, it is demonstrated that the combination of data components that yield the best performance, in a relative sense, can be determined through the diversity-based selection strategy.",
keywords = "Classifier fusion, Data fusion, Diversity, Ensemble classifiers, Event-related potentials",
author = "Lalit Gupta and Srinivas Kota and Molfese, {Dennis L.} and Ravi Vaidyanathan",
year = "2013",
month = "6",
day = "1",
doi = "10.1177/0954411913480621",
language = "English (US)",
volume = "227",
pages = "655--662",
journal = "Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine",
issn = "0954-4119",
publisher = "SAGE Publications Ltd",
number = "6",

}

TY - JOUR

T1 - Pairwise diversity ranking of polychotomous features for ensemble physiological signal classifiers

AU - Gupta, Lalit

AU - Kota, Srinivas

AU - Molfese, Dennis L.

AU - Vaidyanathan, Ravi

PY - 2013/6/1

Y1 - 2013/6/1

N2 - It is well known that fusion classifiers for physiological signal classification with diverse components (classifiers or data sets) outperform those with less diverse components. Determining component diversity, therefore, is of the utmost importance in the design of fusion classifiers that are often employed in clinical diagnostic and numerous other pattern recognition problems. In this article, a new pairwise diversity-based ranking strategy is introduced to select a subset of ensemble components, which when combined will be more diverse than any other component subset of the same size. The strategy is unified in the sense that the components can be classifiers or data sets. Moreover, the classifiers and data sets can be polychotomous. Classifier-fusion and data-fusion systems are formulated based on the diversity-based selection strategy, and the application of the two fusion strategies are demonstrated through the classification of multichannel event-related potentials. It is observed that for both classifier and data fusion, the classification accuracy tends to increase/decrease when the diversity of the component ensemble increases/decreases. For the four sets of 14-channel event-related potentials considered, it is shown that data fusion outperforms classifier fusion. Furthermore, it is demonstrated that the combination of data components that yield the best performance, in a relative sense, can be determined through the diversity-based selection strategy.

AB - It is well known that fusion classifiers for physiological signal classification with diverse components (classifiers or data sets) outperform those with less diverse components. Determining component diversity, therefore, is of the utmost importance in the design of fusion classifiers that are often employed in clinical diagnostic and numerous other pattern recognition problems. In this article, a new pairwise diversity-based ranking strategy is introduced to select a subset of ensemble components, which when combined will be more diverse than any other component subset of the same size. The strategy is unified in the sense that the components can be classifiers or data sets. Moreover, the classifiers and data sets can be polychotomous. Classifier-fusion and data-fusion systems are formulated based on the diversity-based selection strategy, and the application of the two fusion strategies are demonstrated through the classification of multichannel event-related potentials. It is observed that for both classifier and data fusion, the classification accuracy tends to increase/decrease when the diversity of the component ensemble increases/decreases. For the four sets of 14-channel event-related potentials considered, it is shown that data fusion outperforms classifier fusion. Furthermore, it is demonstrated that the combination of data components that yield the best performance, in a relative sense, can be determined through the diversity-based selection strategy.

KW - Classifier fusion

KW - Data fusion

KW - Diversity

KW - Ensemble classifiers

KW - Event-related potentials

UR - http://www.scopus.com/inward/record.url?scp=84884559873&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884559873&partnerID=8YFLogxK

U2 - 10.1177/0954411913480621

DO - 10.1177/0954411913480621

M3 - Article

C2 - 23636746

AN - SCOPUS:84884559873

VL - 227

SP - 655

EP - 662

JO - Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine

JF - Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine

SN - 0954-4119

IS - 6

ER -