A disagreement based co-active learning method for sleep stage classification

Ayşe Betül Yüce; Yusuf Yaslan

doi:10.1109/IWSSIP.2016.7502740

A disagreement based co-active learning method for sleep stage classification

Ayşe Betül Yüce, Yusuf Yaslan

Department of Computer Engineering

Istanbul Technical University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

In this paper a co-active learning framework that uses two feature views for sleep stage classification is proposed. At the beginning of active learning, classifiers are trained on two separate feature views namely Empirical Mode Decomposition (EMD) and frequency domain based energy features which are obtained from different patients' labelled electroencephalography (EEG) data. These classifiers are updated on new patients' unlabeled data using a disagreement based co-active learning framework. Experimental results obtained on Sleep-EDF database show that the proposed co-active learning method outperforms single view based active learning and the proposed method boosts the classification accuracy to approximately 84.85% on average.

Original language	English
Title of host publication	IWSSIP 2016 - Proceedings of the 23rd International Conference on Systems, Signals and Image Processing
Editors	Renata Rybarova, Gregor Rozinaj, Ivan Minarik, Peter Truchly
Publisher	IEEE Computer Society
ISBN (Electronic)	9781467395557
DOIs	https://doi.org/10.1109/IWSSIP.2016.7502740
Publication status	Published - 30 Jun 2016
Event	23rd International Conference on Systems, Signals and Image Processing, IWSSIP 2016 - Bratislava, Slovakia Duration: 23 May 2016 → 25 May 2016

Publication series

Name	International Conference on Systems, Signals, and Image Processing
Volume	2016-June
ISSN (Print)	2157-8672
ISSN (Electronic)	2157-8702

Conference

Conference	23rd International Conference on Systems, Signals and Image Processing, IWSSIP 2016
Country/Territory	Slovakia
City	Bratislava
Period	23/05/16 → 25/05/16

Bibliographical note

Publisher Copyright:
© 2016 IEEE.

Keywords

Active Learning
Empirical Mode Decomposition
Fast Fourier Transform
Sleep Stage Classification

Access to Document

10.1109/IWSSIP.2016.7502740

Cite this

Betül Yüce, A., & Yaslan, Y. (2016). A disagreement based co-active learning method for sleep stage classification. In R. Rybarova, G. Rozinaj, I. Minarik, & P. Truchly (Eds.), IWSSIP 2016 - Proceedings of the 23rd International Conference on Systems, Signals and Image Processing Article 7502740 (International Conference on Systems, Signals, and Image Processing; Vol. 2016-June). IEEE Computer Society. https://doi.org/10.1109/IWSSIP.2016.7502740

Betül Yüce, Ayşe ; Yaslan, Yusuf. / A disagreement based co-active learning method for sleep stage classification. IWSSIP 2016 - Proceedings of the 23rd International Conference on Systems, Signals and Image Processing. editor / Renata Rybarova ; Gregor Rozinaj ; Ivan Minarik ; Peter Truchly. IEEE Computer Society, 2016. (International Conference on Systems, Signals, and Image Processing).

@inproceedings{85b69fcc2631438d84ca0fcc1e3f631b,

title = "A disagreement based co-active learning method for sleep stage classification",

abstract = "In this paper a co-active learning framework that uses two feature views for sleep stage classification is proposed. At the beginning of active learning, classifiers are trained on two separate feature views namely Empirical Mode Decomposition (EMD) and frequency domain based energy features which are obtained from different patients' labelled electroencephalography (EEG) data. These classifiers are updated on new patients' unlabeled data using a disagreement based co-active learning framework. Experimental results obtained on Sleep-EDF database show that the proposed co-active learning method outperforms single view based active learning and the proposed method boosts the classification accuracy to approximately 84.85% on average.",

keywords = "Active Learning, Empirical Mode Decomposition, Fast Fourier Transform, Sleep Stage Classification",

author = "{Bet{\"u}l Y{\"u}ce}, Ay{\c s}e and Yusuf Yaslan",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 23rd International Conference on Systems, Signals and Image Processing, IWSSIP 2016 ; Conference date: 23-05-2016 Through 25-05-2016",

year = "2016",

month = jun,

day = "30",

doi = "10.1109/IWSSIP.2016.7502740",

language = "English",

series = "International Conference on Systems, Signals, and Image Processing",

publisher = "IEEE Computer Society",

editor = "Renata Rybarova and Gregor Rozinaj and Ivan Minarik and Peter Truchly",

booktitle = "IWSSIP 2016 - Proceedings of the 23rd International Conference on Systems, Signals and Image Processing",

address = "United States",

}

Betül Yüce, A & Yaslan, Y 2016, A disagreement based co-active learning method for sleep stage classification. in R Rybarova, G Rozinaj, I Minarik & P Truchly (eds), IWSSIP 2016 - Proceedings of the 23rd International Conference on Systems, Signals and Image Processing., 7502740, International Conference on Systems, Signals, and Image Processing, vol. 2016-June, IEEE Computer Society, 23rd International Conference on Systems, Signals and Image Processing, IWSSIP 2016, Bratislava, Slovakia, 23/05/16. https://doi.org/10.1109/IWSSIP.2016.7502740

A disagreement based co-active learning method for sleep stage classification. / Betül Yüce, Ayşe; Yaslan, Yusuf.
IWSSIP 2016 - Proceedings of the 23rd International Conference on Systems, Signals and Image Processing. ed. / Renata Rybarova; Gregor Rozinaj; Ivan Minarik; Peter Truchly. IEEE Computer Society, 2016. 7502740 (International Conference on Systems, Signals, and Image Processing; Vol. 2016-June).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A disagreement based co-active learning method for sleep stage classification

AU - Betül Yüce, Ayşe

AU - Yaslan, Yusuf

PY - 2016/6/30

Y1 - 2016/6/30

N2 - In this paper a co-active learning framework that uses two feature views for sleep stage classification is proposed. At the beginning of active learning, classifiers are trained on two separate feature views namely Empirical Mode Decomposition (EMD) and frequency domain based energy features which are obtained from different patients' labelled electroencephalography (EEG) data. These classifiers are updated on new patients' unlabeled data using a disagreement based co-active learning framework. Experimental results obtained on Sleep-EDF database show that the proposed co-active learning method outperforms single view based active learning and the proposed method boosts the classification accuracy to approximately 84.85% on average.

AB - In this paper a co-active learning framework that uses two feature views for sleep stage classification is proposed. At the beginning of active learning, classifiers are trained on two separate feature views namely Empirical Mode Decomposition (EMD) and frequency domain based energy features which are obtained from different patients' labelled electroencephalography (EEG) data. These classifiers are updated on new patients' unlabeled data using a disagreement based co-active learning framework. Experimental results obtained on Sleep-EDF database show that the proposed co-active learning method outperforms single view based active learning and the proposed method boosts the classification accuracy to approximately 84.85% on average.

KW - Active Learning

KW - Empirical Mode Decomposition

KW - Fast Fourier Transform

KW - Sleep Stage Classification

UR - http://www.scopus.com/inward/record.url?scp=84978807226&partnerID=8YFLogxK

U2 - 10.1109/IWSSIP.2016.7502740

DO - 10.1109/IWSSIP.2016.7502740

M3 - Conference contribution

AN - SCOPUS:84978807226

T3 - International Conference on Systems, Signals, and Image Processing

BT - IWSSIP 2016 - Proceedings of the 23rd International Conference on Systems, Signals and Image Processing

A2 - Rybarova, Renata

A2 - Rozinaj, Gregor

A2 - Minarik, Ivan

A2 - Truchly, Peter

PB - IEEE Computer Society

T2 - 23rd International Conference on Systems, Signals and Image Processing, IWSSIP 2016

Y2 - 23 May 2016 through 25 May 2016

ER -

Betül Yüce A, Yaslan Y. A disagreement based co-active learning method for sleep stage classification. In Rybarova R, Rozinaj G, Minarik I, Truchly P, editors, IWSSIP 2016 - Proceedings of the 23rd International Conference on Systems, Signals and Image Processing. IEEE Computer Society. 2016. 7502740. (International Conference on Systems, Signals, and Image Processing). doi: 10.1109/IWSSIP.2016.7502740

A disagreement based co-active learning method for sleep stage classification

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this