Hybrid Models for Sentence Readability Assessment

Fengkai Liu; John S. Y. Lee

doi:10.18653/v1/2023.bea-1.37

Hybrid Models for Sentence Readability Assessment

Department of Linguistics and Translation

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

8 Citations (Scopus)

Abstract

Automatic readability assessment (ARA) predicts how difficult it is for the reader to understand a text. While ARA has traditionally been performed at the passage level, there has been increasing interest in ARA at the sentence level, given its applications in downstream tasks such as text simplification and language exercise generation. Recent research has suggested the effectiveness of hybrid approaches for ARA, but they have yet to be applied on the sentence level. We present the first study that compares neural and hybrid models for sentence-level ARA. We conducted experiments on graded sentences from the Wall Street Journal (WSJ) and a dataset derived from the OneStopEnglish corpus. Experimental results show that both neural and hybrid models outperform traditional classifiers trained on linguistic features. Hybrid models obtained the best accuracy on both datasets, surpassing the previous best result reported on the WSJ dataset by almost 13% absolute.

© 2023 Association for Computational Linguistics

Original language	English
Title of host publication	Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023)
Publisher	Association for Computational Linguistics
Pages	448-454
DOIs	https://doi.org/10.18653/v1/2023.bea-1.37
Publication status	Published - 13 Jul 2023
Event	18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023) - Hybrid, Toronto, Canada Duration: 13 Jul 2023 → 13 Jul 2023 https://sig-edu.org/bea/2023

Publication series

Name	Proceedings of the Annual Meeting of the Association for Computational Linguistics
ISSN (Print)	0736-587X

Conference

Conference	18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023)
Country/Territory	Canada
City	Toronto
Period	13/07/23 → 13/07/23
Internet address	https://sig-edu.org/bea/2023

Access Output

10.18653/v1/2023.bea-1.37

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Liu, F., & Lee, J. S. Y. (2023). Hybrid Models for Sentence Readability Assessment. In Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023) (pp. 448-454). (Proceedings of the Annual Meeting of the Association for Computational Linguistics). Association for Computational Linguistics. https://doi.org/10.18653/v1/2023.bea-1.37

@inproceedings{d47cad3b9cd24055b8a2c50451833c70,

title = "Hybrid Models for Sentence Readability Assessment",

abstract = "Automatic readability assessment (ARA) predicts how difficult it is for the reader to understand a text. While ARA has traditionally been performed at the passage level, there has been increasing interest in ARA at the sentence level, given its applications in downstream tasks such as text simplification and language exercise generation. Recent research has suggested the effectiveness of hybrid approaches for ARA, but they have yet to be applied on the sentence level. We present the first study that compares neural and hybrid models for sentence-level ARA. We conducted experiments on graded sentences from the Wall Street Journal (WSJ) and a dataset derived from the OneStopEnglish corpus. Experimental results show that both neural and hybrid models outperform traditional classifiers trained on linguistic features. Hybrid models obtained the best accuracy on both datasets, surpassing the previous best result reported on the WSJ dataset by almost 13% absolute.{\textcopyright} 2023 Association for Computational Linguistics",

author = "Fengkai Liu and Lee, {John S. Y.}",

year = "2023",

month = jul,

day = "13",

doi = "10.18653/v1/2023.bea-1.37",

language = "English",

series = "Proceedings of the Annual Meeting of the Association for Computational Linguistics",

publisher = "Association for Computational Linguistics",

pages = "448--454",

booktitle = "Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023)",

address = "United States",

note = "18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023) ; Conference date: 13-07-2023 Through 13-07-2023",

url = "https://sig-edu.org/bea/2023",

}

Liu, F & Lee, JSY 2023, Hybrid Models for Sentence Readability Assessment. in Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023). Proceedings of the Annual Meeting of the Association for Computational Linguistics, Association for Computational Linguistics, pp. 448-454, 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023), Toronto, Canada, 13/07/23. https://doi.org/10.18653/v1/2023.bea-1.37

Hybrid Models for Sentence Readability Assessment. / Liu, Fengkai ; Lee, John S. Y.
Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023). Association for Computational Linguistics, 2023. p. 448-454 (Proceedings of the Annual Meeting of the Association for Computational Linguistics).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Hybrid Models for Sentence Readability Assessment

AU - Liu, Fengkai

AU - Lee, John S. Y.

PY - 2023/7/13

Y1 - 2023/7/13

N2 - Automatic readability assessment (ARA) predicts how difficult it is for the reader to understand a text. While ARA has traditionally been performed at the passage level, there has been increasing interest in ARA at the sentence level, given its applications in downstream tasks such as text simplification and language exercise generation. Recent research has suggested the effectiveness of hybrid approaches for ARA, but they have yet to be applied on the sentence level. We present the first study that compares neural and hybrid models for sentence-level ARA. We conducted experiments on graded sentences from the Wall Street Journal (WSJ) and a dataset derived from the OneStopEnglish corpus. Experimental results show that both neural and hybrid models outperform traditional classifiers trained on linguistic features. Hybrid models obtained the best accuracy on both datasets, surpassing the previous best result reported on the WSJ dataset by almost 13% absolute.© 2023 Association for Computational Linguistics

AB - Automatic readability assessment (ARA) predicts how difficult it is for the reader to understand a text. While ARA has traditionally been performed at the passage level, there has been increasing interest in ARA at the sentence level, given its applications in downstream tasks such as text simplification and language exercise generation. Recent research has suggested the effectiveness of hybrid approaches for ARA, but they have yet to be applied on the sentence level. We present the first study that compares neural and hybrid models for sentence-level ARA. We conducted experiments on graded sentences from the Wall Street Journal (WSJ) and a dataset derived from the OneStopEnglish corpus. Experimental results show that both neural and hybrid models outperform traditional classifiers trained on linguistic features. Hybrid models obtained the best accuracy on both datasets, surpassing the previous best result reported on the WSJ dataset by almost 13% absolute.© 2023 Association for Computational Linguistics

UR - http://www.scopus.com/inward/record.url?scp=85174544680&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85174544680&origin=recordpage

U2 - 10.18653/v1/2023.bea-1.37

DO - 10.18653/v1/2023.bea-1.37

M3 - RGC 32 - Refereed conference paper (with host publication)

T3 - Proceedings of the Annual Meeting of the Association for Computational Linguistics

SP - 448

EP - 454

BT - Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023)

PB - Association for Computational Linguistics

T2 - 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023)

Y2 - 13 July 2023 through 13 July 2023

ER -

Hybrid Models for Sentence Readability Assessment

Abstract

Publication series

Conference

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this