“Geen makkie”: Interpretable Classification and Simplification of Dutch Text Complexity

Eliza Hobo; Charlotte Pouw; Lisa Beinborn

doi:10.18653/v1/2023.bea-1.42

“Geen makkie”: Interpretable Classification and Simplification of Dutch Text Complexity

Eliza Hobo, Charlotte Pouw, Lisa Beinborn

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

Abstract

An inclusive society needs to facilitate access to information for all of its members, including citizens with low literacy and with non-native language skills. We present an approach to assess Dutch text complexity on the sentence level and conduct an interpretability analysis to explore the link between neural models and linguistic complexity features.¹ Building on these findings, we develop the first contextual lexical simplification model for Dutch and publish a pilot dataset for evaluation. We go beyond previous work which primarily targeted lexical substitution and propose strategies for adjusting the model’s linguistic register to generate simpler candidates. Our results indicate that continual pre-training and multi-task learning with conceptually related tasks are promising directions for ensuring the simplicity of the generated substitutions. Our code repository and the simplification dataset are available on GitHub.

Original language	English
Title of host publication	Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023)
Editors	Ekaterina Kochmar, Jill Burstein, Andrea Horbach, Ronja Laarmann-Quante, Nitin Madnani, Anais Tack, Victoria Yaneva, Zheng Yuan, Torsten Zesch
Publisher	Association for Computational Linguistics (ACL)
Pages	503-517
Number of pages	15
ISBN (Electronic)	9781959429807
DOIs	https://doi.org/10.18653/v1/2023.bea-1.42
Publication status	Published - 2023
Event	18th Workshop on Innovative Use of NLP for Building Educational Applications, BEA 2023 - Toronto, Canada Duration: 13 Jul 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 → …

Bibliographical note

Publisher Copyright:
© 2023 Association for Computational Linguistics.

Funding

Eliza Hobo’s simplification experiments were initiated during an internship at the Gemeente of Amsterdam. Iva Gornishka has been a valuable source of insight and support in this process. Charlotte Pouw’s experiments on readability were initiated in a joint project with Florian Kunneman and Bruna Guedes supported by the Network Institute (VU Amsterdam) through the Academy Assistants Program. Lisa Beinborn’s work was supported by the Dutch National Science Organisation (NWO) through the projects CLARIAHPLUS (CP-W6-19-005) and VENI (Vl.Veni.211C.039).

Funders	Funder number
Dutch National Science organisation
Network Institute
Nederlandse Organisatie voor Wetenschappelijk Onderzoek	CP-W6-19-005
Nederlandse Organisatie voor Wetenschappelijk Onderzoek

Access to Document

10.18653/v1/2023.bea-1.42Licence: CC BY

https://aclanthology.org/2023.bea-1.42/

Persistent URL (handle)

Copy link address

Cite this

Hobo, E., Pouw, C., & Beinborn, L. (2023). “Geen makkie”: Interpretable Classification and Simplification of Dutch Text Complexity. In E. Kochmar, J. Burstein, A. Horbach, R. Laarmann-Quante, N. Madnani, A. Tack, V. Yaneva, Z. Yuan, & T. Zesch (Eds.), Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023) (pp. 503-517). (Proceedings of the Annual Meeting of the Association for Computational Linguistics). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2023.bea-1.42

Hobo, Eliza ; Pouw, Charlotte ; Beinborn, Lisa. / “Geen makkie”: Interpretable Classification and Simplification of Dutch Text Complexity. Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023). editor / Ekaterina Kochmar ; Jill Burstein ; Andrea Horbach ; Ronja Laarmann-Quante ; Nitin Madnani ; Anais Tack ; Victoria Yaneva ; Zheng Yuan ; Torsten Zesch. Association for Computational Linguistics (ACL), 2023. pp. 503-517 (Proceedings of the Annual Meeting of the Association for Computational Linguistics).

@inproceedings{f8b6aa89176b42dcba29130654697004,

title = "“Geen makkie”: Interpretable Classification and Simplification of Dutch Text Complexity",

abstract = "An inclusive society needs to facilitate access to information for all of its members, including citizens with low literacy and with non-native language skills. We present an approach to assess Dutch text complexity on the sentence level and conduct an interpretability analysis to explore the link between neural models and linguistic complexity features.1 Building on these findings, we develop the first contextual lexical simplification model for Dutch and publish a pilot dataset for evaluation. We go beyond previous work which primarily targeted lexical substitution and propose strategies for adjusting the model{\textquoteright}s linguistic register to generate simpler candidates. Our results indicate that continual pre-training and multi-task learning with conceptually related tasks are promising directions for ensuring the simplicity of the generated substitutions. Our code repository and the simplification dataset are available on GitHub.",

author = "Eliza Hobo and Charlotte Pouw and Lisa Beinborn",

note = "Publisher Copyright: {\textcopyright} 2023 Association for Computational Linguistics.; 18th Workshop on Innovative Use of NLP for Building Educational Applications, BEA 2023 ; Conference date: 13-07-2023",

year = "2023",

doi = "10.18653/v1/2023.bea-1.42",

language = "English",

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

publisher = "Association for Computational Linguistics (ACL)",

pages = "503--517",

editor = "Ekaterina Kochmar and Jill Burstein and Andrea Horbach and Ronja Laarmann-Quante and Nitin Madnani and Anais Tack and Victoria Yaneva and Zheng Yuan and Torsten Zesch",

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

address = "United States",

}

Hobo, E, Pouw, C & Beinborn, L 2023, “Geen makkie”: Interpretable Classification and Simplification of Dutch Text Complexity. in E Kochmar, J Burstein, A Horbach, R Laarmann-Quante, N Madnani, A Tack, V Yaneva, Z Yuan & T Zesch (eds), 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 (ACL), pp. 503-517, 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-1.42

“Geen makkie”: Interpretable Classification and Simplification of Dutch Text Complexity. / Hobo, Eliza; Pouw, Charlotte; Beinborn, Lisa.
Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023). ed. / Ekaterina Kochmar; Jill Burstein; Andrea Horbach; Ronja Laarmann-Quante; Nitin Madnani; Anais Tack; Victoria Yaneva; Zheng Yuan; Torsten Zesch. Association for Computational Linguistics (ACL), 2023. p. 503-517 (Proceedings of the Annual Meeting of the Association for Computational Linguistics).

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

TY - GEN

T1 - “Geen makkie”: Interpretable Classification and Simplification of Dutch Text Complexity

AU - Hobo, Eliza

AU - Pouw, Charlotte

AU - Beinborn, Lisa

PY - 2023

Y1 - 2023

N2 - An inclusive society needs to facilitate access to information for all of its members, including citizens with low literacy and with non-native language skills. We present an approach to assess Dutch text complexity on the sentence level and conduct an interpretability analysis to explore the link between neural models and linguistic complexity features.1 Building on these findings, we develop the first contextual lexical simplification model for Dutch and publish a pilot dataset for evaluation. We go beyond previous work which primarily targeted lexical substitution and propose strategies for adjusting the model’s linguistic register to generate simpler candidates. Our results indicate that continual pre-training and multi-task learning with conceptually related tasks are promising directions for ensuring the simplicity of the generated substitutions. Our code repository and the simplification dataset are available on GitHub.

AB - An inclusive society needs to facilitate access to information for all of its members, including citizens with low literacy and with non-native language skills. We present an approach to assess Dutch text complexity on the sentence level and conduct an interpretability analysis to explore the link between neural models and linguistic complexity features.1 Building on these findings, we develop the first contextual lexical simplification model for Dutch and publish a pilot dataset for evaluation. We go beyond previous work which primarily targeted lexical substitution and propose strategies for adjusting the model’s linguistic register to generate simpler candidates. Our results indicate that continual pre-training and multi-task learning with conceptually related tasks are promising directions for ensuring the simplicity of the generated substitutions. Our code repository and the simplification dataset are available on GitHub.

UR - http://www.scopus.com/inward/record.url?scp=85174502154&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85174502154&partnerID=8YFLogxK

U2 - 10.18653/v1/2023.bea-1.42

DO - 10.18653/v1/2023.bea-1.42

M3 - Conference contribution

AN - SCOPUS:85174502154

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

SP - 503

EP - 517

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

A2 - Kochmar, Ekaterina

A2 - Burstein, Jill

A2 - Horbach, Andrea

A2 - Laarmann-Quante, Ronja

A2 - Madnani, Nitin

A2 - Tack, Anais

A2 - Yaneva, Victoria

A2 - Yuan, Zheng

A2 - Zesch, Torsten

PB - Association for Computational Linguistics (ACL)

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

Y2 - 13 July 2023

ER -

Hobo E, Pouw C, Beinborn L. “Geen makkie”: Interpretable Classification and Simplification of Dutch Text Complexity. In Kochmar E, Burstein J, Horbach A, Laarmann-Quante R, Madnani N, Tack A, Yaneva V, Yuan Z, Zesch T, editors, Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023). Association for Computational Linguistics (ACL). 2023. p. 503-517. (Proceedings of the Annual Meeting of the Association for Computational Linguistics). Epub 2023 Jul. doi: 10.18653/v1/2023.bea-1.42

“Geen makkie”: Interpretable Classification and Simplification of Dutch Text Complexity

Abstract

Publication series

Conference

Bibliographical note

Funding

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this