Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification

Intan Nurma Yulita; Mohamad Ivan Fanany; Aniati Murni Arymuthy

doi:10.1016/j.procs.2017.10.042

Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification

Intan Nurma Yulita, Mohamad Ivan Fanany, Aniati Murni Arymuthy

Department of Computer Science

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

42 Citations (Scopus)

Abstract

The study examines the use of quantization to be applied to Bi-directional Long Short-Term Memory (Bi-LSTM), a combination of the two called qBi-LSTM. Quantization used comes from Deep Belief Networks (DBN). It selected DBN for its superiority as a generative model of Deep Learning in producing an optimal artificial feature. Development of qBi-LSTM is expected to improve the performance of Bi-LSTM and also provide efficient time. The qBi-LSTM test is applied for sleep stage classification on St. Vincent's University Hospital / University College Dublin's Sleep Apnea Database. The result shows that qBi-LSTM has the highest performance compared to Bi-LSTM and DBN with precision, recall and F-measure values of 86.00%, 72.10%, and 75.27%. The best qBi-LSTM performance is to classify Stage 2 but still fails to classify the stage of REM (Rapid Eye Movement).

Original language	English
Title of host publication	2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017
Editors	Wdodo Budiharto, Dewi Suryani, Lili A. Wulandhari, Andry Chowanda, Alexander A.S. Gunawan, Novita Hanafiah, Hanry Ham, Meiliana
Publisher	Elsevier B.V.
Pages	530-538
Number of pages	9
ISBN (Print)	9781510849914
DOIs	https://doi.org/10.1016/j.procs.2017.10.042
Publication status	Published - 2017
Event	2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017 - Bali, Indonesia Duration: 13 Oct 2017 → 14 Oct 2017

Publication series

Name	Procedia Computer Science
Volume	116
ISSN (Electronic)	1877-0509

Conference

Conference	2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017
Country/Territory	Indonesia
City	Bali
Period	13/10/17 → 14/10/17

Keywords

Bi-directional Long Short-Term Memory
Deep Belief Networks
Long Short-Term Memory
Sleep Stage Classification

Access to Document

10.1016/j.procs.2017.10.042

Cite this

Yulita, I. N., Fanany, M. I., & Arymuthy, A. M. (2017). Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification. In W. Budiharto, D. Suryani, L. A. Wulandhari, A. Chowanda, A. A. S. Gunawan, N. Hanafiah, H. Ham, & Meiliana (Eds.), 2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017 (pp. 530-538). (Procedia Computer Science; Vol. 116). Elsevier B.V.. https://doi.org/10.1016/j.procs.2017.10.042

Yulita, Intan Nurma ; Fanany, Mohamad Ivan ; Arymuthy, Aniati Murni. / Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification. 2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017. editor / Wdodo Budiharto ; Dewi Suryani ; Lili A. Wulandhari ; Andry Chowanda ; Alexander A.S. Gunawan ; Novita Hanafiah ; Hanry Ham ; Meiliana. Elsevier B.V., 2017. pp. 530-538 (Procedia Computer Science).

@inproceedings{4cc1ef6c90114fb6b7ef53fcde278641,

title = "Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification",

abstract = "The study examines the use of quantization to be applied to Bi-directional Long Short-Term Memory (Bi-LSTM), a combination of the two called qBi-LSTM. Quantization used comes from Deep Belief Networks (DBN). It selected DBN for its superiority as a generative model of Deep Learning in producing an optimal artificial feature. Development of qBi-LSTM is expected to improve the performance of Bi-LSTM and also provide efficient time. The qBi-LSTM test is applied for sleep stage classification on St. Vincent's University Hospital / University College Dublin's Sleep Apnea Database. The result shows that qBi-LSTM has the highest performance compared to Bi-LSTM and DBN with precision, recall and F-measure values of 86.00%, 72.10%, and 75.27%. The best qBi-LSTM performance is to classify Stage 2 but still fails to classify the stage of REM (Rapid Eye Movement).",

keywords = "Bi-directional Long Short-Term Memory, Deep Belief Networks, Long Short-Term Memory, Sleep Stage Classification",

author = "Yulita, {Intan Nurma} and Fanany, {Mohamad Ivan} and Arymuthy, {Aniati Murni}",

year = "2017",

doi = "10.1016/j.procs.2017.10.042",

language = "English",

isbn = "9781510849914",

series = "Procedia Computer Science",

publisher = "Elsevier B.V.",

pages = "530--538",

editor = "Wdodo Budiharto and Dewi Suryani and Wulandhari, {Lili A.} and Andry Chowanda and Gunawan, {Alexander A.S.} and Novita Hanafiah and Hanry Ham and Meiliana",

booktitle = "2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017",

address = "Netherlands",

note = "2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017 ; Conference date: 13-10-2017 Through 14-10-2017",

}

Yulita, IN, Fanany, MI & Arymuthy, AM 2017, Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification. in W Budiharto, D Suryani, LA Wulandhari, A Chowanda, AAS Gunawan, N Hanafiah, H Ham & Meiliana (eds), 2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017. Procedia Computer Science, vol. 116, Elsevier B.V., pp. 530-538, 2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017, Bali, Indonesia, 13/10/17. https://doi.org/10.1016/j.procs.2017.10.042

Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification. / Yulita, Intan Nurma; Fanany, Mohamad Ivan; Arymuthy, Aniati Murni.
2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017. ed. / Wdodo Budiharto; Dewi Suryani; Lili A. Wulandhari; Andry Chowanda; Alexander A.S. Gunawan; Novita Hanafiah; Hanry Ham; Meiliana. Elsevier B.V., 2017. p. 530-538 (Procedia Computer Science; Vol. 116).

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

TY - GEN

T1 - Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification

AU - Yulita, Intan Nurma

AU - Fanany, Mohamad Ivan

AU - Arymuthy, Aniati Murni

PY - 2017

Y1 - 2017

N2 - The study examines the use of quantization to be applied to Bi-directional Long Short-Term Memory (Bi-LSTM), a combination of the two called qBi-LSTM. Quantization used comes from Deep Belief Networks (DBN). It selected DBN for its superiority as a generative model of Deep Learning in producing an optimal artificial feature. Development of qBi-LSTM is expected to improve the performance of Bi-LSTM and also provide efficient time. The qBi-LSTM test is applied for sleep stage classification on St. Vincent's University Hospital / University College Dublin's Sleep Apnea Database. The result shows that qBi-LSTM has the highest performance compared to Bi-LSTM and DBN with precision, recall and F-measure values of 86.00%, 72.10%, and 75.27%. The best qBi-LSTM performance is to classify Stage 2 but still fails to classify the stage of REM (Rapid Eye Movement).

AB - The study examines the use of quantization to be applied to Bi-directional Long Short-Term Memory (Bi-LSTM), a combination of the two called qBi-LSTM. Quantization used comes from Deep Belief Networks (DBN). It selected DBN for its superiority as a generative model of Deep Learning in producing an optimal artificial feature. Development of qBi-LSTM is expected to improve the performance of Bi-LSTM and also provide efficient time. The qBi-LSTM test is applied for sleep stage classification on St. Vincent's University Hospital / University College Dublin's Sleep Apnea Database. The result shows that qBi-LSTM has the highest performance compared to Bi-LSTM and DBN with precision, recall and F-measure values of 86.00%, 72.10%, and 75.27%. The best qBi-LSTM performance is to classify Stage 2 but still fails to classify the stage of REM (Rapid Eye Movement).

KW - Bi-directional Long Short-Term Memory

KW - Deep Belief Networks

KW - Long Short-Term Memory

KW - Sleep Stage Classification

UR - http://www.scopus.com/inward/record.url?scp=85040016983&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2017.10.042

DO - 10.1016/j.procs.2017.10.042

M3 - Conference contribution

AN - SCOPUS:85040016983

SN - 9781510849914

T3 - Procedia Computer Science

SP - 530

EP - 538

BT - 2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017

A2 - Budiharto, Wdodo

A2 - Suryani, Dewi

A2 - Wulandhari, Lili A.

A2 - Chowanda, Andry

A2 - Gunawan, Alexander A.S.

A2 - Hanafiah, Novita

A2 - Ham, Hanry

A2 - Meiliana, null

PB - Elsevier B.V.

T2 - 2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017

Y2 - 13 October 2017 through 14 October 2017

ER -

Yulita IN, Fanany MI, Arymuthy AM. Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification. In Budiharto W, Suryani D, Wulandhari LA, Chowanda A, Gunawan AAS, Hanafiah N, Ham H, Meiliana, editors, 2nd International Conference on Computer Science and Computational Intelligence, ICCSCI 2017. Elsevier B.V. 2017. p. 530-538. (Procedia Computer Science). doi: 10.1016/j.procs.2017.10.042

Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this