Comparison of automatic extraction of research highlights and abstracts of journal articles

Wai Ming WANG*, Eric Wing Kuen SEE-TO, Hong Tao LIN, Zhi LI

*Corresponding author for this work

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

Abstract

Many academic journals1 require authors to submit a list of bullet points, called highlights. Highlights provide concise information for readers to review the article quickly. However, some journals and old articles do not have the highlights. It is useful to develop an automated method for extracting highlights of these articles. In this paper, we study the important features of research highlight extraction and identify the differences between extraction of highlights and abstracts of journal articles. We use information science related journal articles as test data. We quantitatively evaluate 23 common unsupervised extractive text summarization methods. The results show that the application of extractive text summarization is suitable for research highlight extraction. In particular, TextRank obtains the highest recall and the title word method achieves the highest precision. The result could help researchers developing new methods of highlight extraction.

Original languageEnglish
Title of host publicationProceedings of 2nd International Conference on Computer Science and Application Engineering, CSAE 2018
EditorsAli Emrouznejad
PublisherAssociation for Computing Machinery
ISBN (Electronic)9781450365123
DOIs
Publication statusPublished - 22 Oct 2018
Event2nd International Conference on Computer Science and Application Engineering, CSAE 2018 - Hohhot, China
Duration: 22 Oct 201824 Oct 2018

Conference

Conference2nd International Conference on Computer Science and Application Engineering, CSAE 2018
CountryChina
CityHohhot
Period22/10/1824/10/18

Fingerprint

Information science

Keywords

  • Abstract extraction
  • Automatic text summarization
  • Extractive-based summarization
  • Research highlight extraction

Cite this

WANG, W. M., SEE-TO, E. W. K., LIN, H. T., & LI, Z. (2018). Comparison of automatic extraction of research highlights and abstracts of journal articles. In A. Emrouznejad (Ed.), Proceedings of 2nd International Conference on Computer Science and Application Engineering, CSAE 2018 [a132] Association for Computing Machinery. https://doi.org/10.1145/3207677.3277979
WANG, Wai Ming ; SEE-TO, Eric Wing Kuen ; LIN, Hong Tao ; LI, Zhi. / Comparison of automatic extraction of research highlights and abstracts of journal articles. Proceedings of 2nd International Conference on Computer Science and Application Engineering, CSAE 2018. editor / Ali Emrouznejad. Association for Computing Machinery, 2018.
@inproceedings{6b3adb7bcb8741468d353addd8a889cc,
title = "Comparison of automatic extraction of research highlights and abstracts of journal articles",
abstract = "Many academic journals1 require authors to submit a list of bullet points, called highlights. Highlights provide concise information for readers to review the article quickly. However, some journals and old articles do not have the highlights. It is useful to develop an automated method for extracting highlights of these articles. In this paper, we study the important features of research highlight extraction and identify the differences between extraction of highlights and abstracts of journal articles. We use information science related journal articles as test data. We quantitatively evaluate 23 common unsupervised extractive text summarization methods. The results show that the application of extractive text summarization is suitable for research highlight extraction. In particular, TextRank obtains the highest recall and the title word method achieves the highest precision. The result could help researchers developing new methods of highlight extraction.",
keywords = "Abstract extraction, Automatic text summarization, Extractive-based summarization, Research highlight extraction",
author = "WANG, {Wai Ming} and SEE-TO, {Eric Wing Kuen} and LIN, {Hong Tao} and Zhi LI",
year = "2018",
month = "10",
day = "22",
doi = "10.1145/3207677.3277979",
language = "English",
editor = "Ali Emrouznejad",
booktitle = "Proceedings of 2nd International Conference on Computer Science and Application Engineering, CSAE 2018",
publisher = "Association for Computing Machinery",

}

WANG, WM, SEE-TO, EWK, LIN, HT & LI, Z 2018, Comparison of automatic extraction of research highlights and abstracts of journal articles. in A Emrouznejad (ed.), Proceedings of 2nd International Conference on Computer Science and Application Engineering, CSAE 2018., a132, Association for Computing Machinery, 2nd International Conference on Computer Science and Application Engineering, CSAE 2018, Hohhot, China, 22/10/18. https://doi.org/10.1145/3207677.3277979

Comparison of automatic extraction of research highlights and abstracts of journal articles. / WANG, Wai Ming; SEE-TO, Eric Wing Kuen; LIN, Hong Tao; LI, Zhi.

Proceedings of 2nd International Conference on Computer Science and Application Engineering, CSAE 2018. ed. / Ali Emrouznejad. Association for Computing Machinery, 2018. a132.

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

TY - GEN

T1 - Comparison of automatic extraction of research highlights and abstracts of journal articles

AU - WANG, Wai Ming

AU - SEE-TO, Eric Wing Kuen

AU - LIN, Hong Tao

AU - LI, Zhi

PY - 2018/10/22

Y1 - 2018/10/22

N2 - Many academic journals1 require authors to submit a list of bullet points, called highlights. Highlights provide concise information for readers to review the article quickly. However, some journals and old articles do not have the highlights. It is useful to develop an automated method for extracting highlights of these articles. In this paper, we study the important features of research highlight extraction and identify the differences between extraction of highlights and abstracts of journal articles. We use information science related journal articles as test data. We quantitatively evaluate 23 common unsupervised extractive text summarization methods. The results show that the application of extractive text summarization is suitable for research highlight extraction. In particular, TextRank obtains the highest recall and the title word method achieves the highest precision. The result could help researchers developing new methods of highlight extraction.

AB - Many academic journals1 require authors to submit a list of bullet points, called highlights. Highlights provide concise information for readers to review the article quickly. However, some journals and old articles do not have the highlights. It is useful to develop an automated method for extracting highlights of these articles. In this paper, we study the important features of research highlight extraction and identify the differences between extraction of highlights and abstracts of journal articles. We use information science related journal articles as test data. We quantitatively evaluate 23 common unsupervised extractive text summarization methods. The results show that the application of extractive text summarization is suitable for research highlight extraction. In particular, TextRank obtains the highest recall and the title word method achieves the highest precision. The result could help researchers developing new methods of highlight extraction.

KW - Abstract extraction

KW - Automatic text summarization

KW - Extractive-based summarization

KW - Research highlight extraction

UR - http://www.scopus.com/inward/record.url?scp=85056888482&partnerID=8YFLogxK

U2 - 10.1145/3207677.3277979

DO - 10.1145/3207677.3277979

M3 - Conference paper (refereed)

BT - Proceedings of 2nd International Conference on Computer Science and Application Engineering, CSAE 2018

A2 - Emrouznejad, Ali

PB - Association for Computing Machinery

ER -

WANG WM, SEE-TO EWK, LIN HT, LI Z. Comparison of automatic extraction of research highlights and abstracts of journal articles. In Emrouznejad A, editor, Proceedings of 2nd International Conference on Computer Science and Application Engineering, CSAE 2018. Association for Computing Machinery. 2018. a132 https://doi.org/10.1145/3207677.3277979