Biomedical text recognition using convolutional neural networks: Content based deep learning

Sisir Joshi; Abeer Alsadoon; S. M.N.Arosha Senanayake; P. W.C. Prasad; Abdul Ghani Naim; Amr Elchouemi

doi:10.1007/978-3-030-63119-2_48

Biomedical text recognition using convolutional neural networks: Content based deep learning

Sisir Joshi, Abeer Alsadoon, S. M.N.Arosha Senanayake, P. W.C. Prasad, Abdul Ghani Naim, Amr Elchouemi

Research output: Book chapter/Published conference paper › Conference paper › peer-review

Abstract

Named Entity Recognition (NER) targets to automatically detect the drug and disease mentions from biomedical texts and is fundamental step in the biomedical text mining. Although deep learning has been successfully implemented, the accuracy and processing time are still major issues preventing it from achieving NMR. This research aims to upgrade the accuracy of classification while decreasing the processing time, by paying more attention to significant areas of NMR. The novel proposed system consists of a Bi-Directional Long Short-Term Memory with Conditional Random Field (BiLSTM-CRF) using dropout strategy to effectively prevent overfitting and enhancing the generalization abilities. The system built includes the attention mechanism and attention fusion for redistributing the weight of samples belonging to each class in order to compensate the problem occurring from data imbalance and to focus only on the critical areas of the observed things and ignoring non-critical areas.

Original language	English
Title of host publication	Advances in Computational Collective Intelligence
Subtitle of host publication	12th International Conference, ICCCI 2020, Proceedings
Editors	Marcin Hernes, Krystian Wojtkiewicz, Edward Szczerbicki
Place of Publication	Switzerland
Publisher	Springer
Pages	593-604
Number of pages	12
ISBN (Electronic)	9783030631192
ISBN (Print)	9783030631185
DOIs	https://doi.org/10.1007/978-3-030-63119-2_48
Publication status	E-pub ahead of print - 19 Nov 2020
Event	12th International Conference on Computational Collective Intelligence, ICCCI 2020 - Online, Da Nang, Viet Nam Duration: 30 Nov 2020 → 03 Dec 2020 http://fit.ntt.edu.vn/-12th-international-conference-on-computational-collective-intelligence-224/ https://iccci.pwr.edu.pl/2020/ (Conference website) https://drive.google.com/viewerng/viewer?url=https://iccci.pwr.edu.pl/2020/download/ICCCI_2020_Conference_OnLine_Program_v.7.1.pdf (Program)

Publication series

Name	Communications in Computer and Information Science
Volume	1287
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	12th International Conference on Computational Collective Intelligence, ICCCI 2020
Country/Territory	Viet Nam
City	Da Nang
Period	30/11/20 → 03/12/20
Other	The aim of this conference is to provide an internationally respected forum for scientific research in the computer-based methods of collective intelligence and their applications. The scope of ICCCI 2019 includes all topics related with processing data and knowledge from autonomous sources, especially in: collective processing, knowledge integration, data integration, group decision making, multi-criteria decision making, consensus computing, and social networks. The conference is hosted by The University of Danang, Vietnam - Korea University of Information and Communication Technology, Vietnam and jointly organized by Wrocław University of Science and Technology, Poland and Wrocław University of Economics, Poland, in cooperation with IEEE SMC Technical Committee on Computational Collective Intelligence, European Research Center for Information Systems (ERCIS), and International University – VNU-HCMC, Vietnam.
Internet address	http://fit.ntt.edu.vn/-12th-international-conference-on-computational-collective-intelligence-224/ https://iccci.pwr.edu.pl/2020/ (Conference website) https://drive.google.com/viewerng/viewer?url=https://iccci.pwr.edu.pl/2020/download/ICCCI_2020_Conference_OnLine_Program_v.7.1.pdf (Program)

Access to Document

10.1007/978-3-030-63119-2_48

Cite this

Joshi, S., Alsadoon, A., Senanayake, S. M. N. A., Prasad, P. W. C., Naim, A. G., & Elchouemi, A. (2020). Biomedical text recognition using convolutional neural networks: Content based deep learning. In M. Hernes, K. Wojtkiewicz, & E. Szczerbicki (Eds.), Advances in Computational Collective Intelligence: 12th International Conference, ICCCI 2020, Proceedings (pp. 593-604). (Communications in Computer and Information Science; Vol. 1287). Springer. Advance online publication. https://doi.org/10.1007/978-3-030-63119-2_48

Joshi, Sisir ; Alsadoon, Abeer ; Senanayake, S. M.N.Arosha et al. / Biomedical text recognition using convolutional neural networks : Content based deep learning. Advances in Computational Collective Intelligence: 12th International Conference, ICCCI 2020, Proceedings. editor / Marcin Hernes ; Krystian Wojtkiewicz ; Edward Szczerbicki. Switzerland : Springer, 2020. pp. 593-604 (Communications in Computer and Information Science).

@inproceedings{6252b997f56148328298514c88d25a7c,

title = "Biomedical text recognition using convolutional neural networks: Content based deep learning",

abstract = "Named Entity Recognition (NER) targets to automatically detect the drug and disease mentions from biomedical texts and is fundamental step in the biomedical text mining. Although deep learning has been successfully implemented, the accuracy and processing time are still major issues preventing it from achieving NMR. This research aims to upgrade the accuracy of classification while decreasing the processing time, by paying more attention to significant areas of NMR. The novel proposed system consists of a Bi-Directional Long Short-Term Memory with Conditional Random Field (BiLSTM-CRF) using dropout strategy to effectively prevent overfitting and enhancing the generalization abilities. The system built includes the attention mechanism and attention fusion for redistributing the weight of samples belonging to each class in order to compensate the problem occurring from data imbalance and to focus only on the critical areas of the observed things and ignoring non-critical areas.",

keywords = "Bi-LSTM (Bi-Directional Long Short-Term Memory), Biomedical text, Convolutional neural networks, Named entity recognition, Text mining",

author = "Sisir Joshi and Abeer Alsadoon and Senanayake, {S. M.N.Arosha} and Prasad, {P. W.C.} and Naim, {Abdul Ghani} and Amr Elchouemi",

note = "Includes bibliographical references.; 12th International Conference on Computational Collective Intelligence, ICCCI 2020 ; Conference date: 30-11-2020 Through 03-12-2020",

year = "2020",

month = nov,

day = "19",

doi = "10.1007/978-3-030-63119-2_48",

language = "English",

isbn = "9783030631185",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "593--604",

editor = "Marcin Hernes and Krystian Wojtkiewicz and Edward Szczerbicki",

booktitle = "Advances in Computational Collective Intelligence",

address = "United States",

url = "http://fit.ntt.edu.vn/-12th-international-conference-on-computational-collective-intelligence-224/, https://iccci.pwr.edu.pl/2020/, https://drive.google.com/viewerng/viewer?url=https://iccci.pwr.edu.pl/2020/download/ICCCI_2020_Conference_OnLine_Program_v.7.1.pdf",

}

Joshi, S, Alsadoon, A, Senanayake, SMNA, Prasad, PWC, Naim, AG & Elchouemi, A 2020, Biomedical text recognition using convolutional neural networks: Content based deep learning. in M Hernes, K Wojtkiewicz & E Szczerbicki (eds), Advances in Computational Collective Intelligence: 12th International Conference, ICCCI 2020, Proceedings. Communications in Computer and Information Science, vol. 1287, Springer, Switzerland, pp. 593-604, 12th International Conference on Computational Collective Intelligence, ICCCI 2020, Da Nang, Viet Nam, 30/11/20. https://doi.org/10.1007/978-3-030-63119-2_48

Biomedical text recognition using convolutional neural networks: Content based deep learning. / Joshi, Sisir; Alsadoon, Abeer; Senanayake, S. M.N.Arosha et al.
Advances in Computational Collective Intelligence: 12th International Conference, ICCCI 2020, Proceedings. ed. / Marcin Hernes; Krystian Wojtkiewicz; Edward Szczerbicki. Switzerland: Springer, 2020. p. 593-604 (Communications in Computer and Information Science; Vol. 1287).

Research output: Book chapter/Published conference paper › Conference paper › peer-review

TY - GEN

T1 - Biomedical text recognition using convolutional neural networks

T2 - 12th International Conference on Computational Collective Intelligence, ICCCI 2020

AU - Joshi, Sisir

AU - Alsadoon, Abeer

AU - Senanayake, S. M.N.Arosha

AU - Prasad, P. W.C.

AU - Naim, Abdul Ghani

AU - Elchouemi, Amr

N1 - Includes bibliographical references.

PY - 2020/11/19

Y1 - 2020/11/19

N2 - Named Entity Recognition (NER) targets to automatically detect the drug and disease mentions from biomedical texts and is fundamental step in the biomedical text mining. Although deep learning has been successfully implemented, the accuracy and processing time are still major issues preventing it from achieving NMR. This research aims to upgrade the accuracy of classification while decreasing the processing time, by paying more attention to significant areas of NMR. The novel proposed system consists of a Bi-Directional Long Short-Term Memory with Conditional Random Field (BiLSTM-CRF) using dropout strategy to effectively prevent overfitting and enhancing the generalization abilities. The system built includes the attention mechanism and attention fusion for redistributing the weight of samples belonging to each class in order to compensate the problem occurring from data imbalance and to focus only on the critical areas of the observed things and ignoring non-critical areas.

AB - Named Entity Recognition (NER) targets to automatically detect the drug and disease mentions from biomedical texts and is fundamental step in the biomedical text mining. Although deep learning has been successfully implemented, the accuracy and processing time are still major issues preventing it from achieving NMR. This research aims to upgrade the accuracy of classification while decreasing the processing time, by paying more attention to significant areas of NMR. The novel proposed system consists of a Bi-Directional Long Short-Term Memory with Conditional Random Field (BiLSTM-CRF) using dropout strategy to effectively prevent overfitting and enhancing the generalization abilities. The system built includes the attention mechanism and attention fusion for redistributing the weight of samples belonging to each class in order to compensate the problem occurring from data imbalance and to focus only on the critical areas of the observed things and ignoring non-critical areas.

KW - Bi-LSTM (Bi-Directional Long Short-Term Memory)

KW - Biomedical text

KW - Convolutional neural networks

KW - Named entity recognition

KW - Text mining

UR - http://www.scopus.com/inward/record.url?scp=85097073712&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85097073712&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-63119-2_48

DO - 10.1007/978-3-030-63119-2_48

M3 - Conference paper

AN - SCOPUS:85097073712

SN - 9783030631185

T3 - Communications in Computer and Information Science

SP - 593

EP - 604

BT - Advances in Computational Collective Intelligence

A2 - Hernes, Marcin

A2 - Wojtkiewicz, Krystian

A2 - Szczerbicki, Edward

PB - Springer

CY - Switzerland

Y2 - 30 November 2020 through 3 December 2020

ER -

Joshi S, Alsadoon A, Senanayake SMNA, Prasad PWC, Naim AG, Elchouemi A. Biomedical text recognition using convolutional neural networks: Content based deep learning. In Hernes M, Wojtkiewicz K, Szczerbicki E, editors, Advances in Computational Collective Intelligence: 12th International Conference, ICCCI 2020, Proceedings. Switzerland: Springer. 2020. p. 593-604. (Communications in Computer and Information Science). Epub 2020 Nov 19. doi: 10.1007/978-3-030-63119-2_48

Biomedical text recognition using convolutional neural networks: Content based deep learning

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this