A Multilingual Approach to Identify and Classify Exceptional Measures against COVID-19

Georgios Tziafas; Eugenie De Saint-Phalle; Wietse de Vries; Clara Egger; Tommaso Caselli

A Multilingual Approach to Identify and Classify Exceptional Measures against COVID-19

Georgios Tziafas, Eugenie De Saint-Phalle, Wietse de Vries, Clara Egger, Tommaso Caselli

Center for International Relations Research (CIRR)

Onderzoeksoutput › Academic › peer review

6 Citaten (Scopus)

164 Downloads (Pure)

Samenvatting

The COVID-19 pandemic has witnessed the implementations of exceptional measures by governments across the world to counteract its impact. This work presents the initial results of an on-going project, EXCEPTIUS, aiming to automatically identify, classify and com- pare exceptional measures against COVID-19 across 32 countries in Europe. To this goal, we created a corpus of legal documents with sentence-level annotations of eight different classes of exceptional measures that are im- plemented across these countries. We evalu- ated multiple multi-label classifiers on a manu- ally annotated corpus at sentence level. The XLM-RoBERTa model achieves highest per- formance on this multilingual multi-label clas- sification task, with a macro-average F1 score of 59.8%.

Originele taal-2	English
Titel	Proceedings of the Natural Legal Language Processing Workshop 2021
Redacteuren	Nikolaos Aletras, Ion Androutsopoulos, Leslie Barrett, Catalina Goanta, Daniel Preotiuc-Pietro
Uitgeverij	Association for Computational Linguistics (ACL)
Pagina's	46-62
Aantal pagina's	17
Status	Published - 2021
Evenement	Natural Legal Language Processing Workshop 2021 - Duur: 10-nov.-2021 → … https://nllpw.org/workshop/

Conference

Conference	Natural Legal Language Processing Workshop 2021
Periode	10/11/2021 → …
Internet adres	https://nllpw.org/workshop/

Toegang tot document

A Multilingual Approach to Identify and Classify Exceptional Measures against COVID-19Final publisher's version, 317 KBLicentie: CC BY

Handle.net

Permanente koppeling

EXCEPTIUS Corpus
Caselli, T. (Contributor), Egger, C. (Contributor), Tziafas, G. (Contributor) & De Saint-Phalle, E. (Contributor), DataverseNL, 29-nov.-2021
DOI: 10.34894/zuwaps
Dataset

EXCEPTIUS Exceptional powers in time of Sars-CoV-2 crisis
Egger, C. (Hoofdonderzoeker) & Aarts, K. (CoPI)
07/09/2020 → 06/09/2022
Project: Research

Citeer dit

Tziafas, G., De Saint-Phalle, E., de Vries, W., Egger, C., & Caselli, T. (2021). A Multilingual Approach to Identify and Classify Exceptional Measures against COVID-19. In N. Aletras, I. Androutsopoulos, L. Barrett, C. Goanta, & D. Preotiuc-Pietro (editors), Proceedings of the Natural Legal Language Processing Workshop 2021 (blz. 46-62). Association for Computational Linguistics (ACL).

Tziafas, Georgios ; De Saint-Phalle, Eugenie ; de Vries, Wietse et al. / A Multilingual Approach to Identify and Classify Exceptional Measures against COVID-19. Proceedings of the Natural Legal Language Processing Workshop 2021. redacteur / Nikolaos Aletras ; Ion Androutsopoulos ; Leslie Barrett ; Catalina Goanta ; Daniel Preotiuc-Pietro. Association for Computational Linguistics (ACL), 2021. blz. 46-62

@inproceedings{ebbdc647fdc748c898c6a21212460022,

title = "A Multilingual Approach to Identify and Classify Exceptional Measures against COVID-19",

abstract = "The COVID-19 pandemic has witnessed the implementations of exceptional measures by governments across the world to counteract its impact. This work presents the initial results of an on-going project, EXCEPTIUS, aiming to automatically identify, classify and com- pare exceptional measures against COVID-19 across 32 countries in Europe. To this goal, we created a corpus of legal documents with sentence-level annotations of eight different classes of exceptional measures that are im- plemented across these countries. We evalu- ated multiple multi-label classifiers on a manu- ally annotated corpus at sentence level. The XLM-RoBERTa model achieves highest per- formance on this multilingual multi-label clas- sification task, with a macro-average F1 score of 59.8%.",

author = "Georgios Tziafas and {De Saint-Phalle}, Eugenie and {de Vries}, Wietse and Clara Egger and Tommaso Caselli",

year = "2021",

language = "English",

pages = "46--62",

editor = "Nikolaos Aletras and Ion Androutsopoulos and Leslie Barrett and Catalina Goanta and Daniel Preotiuc-Pietro",

booktitle = "Proceedings of the Natural Legal Language Processing Workshop 2021",

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

note = "Natural Legal Language Processing Workshop 2021 ; Conference date: 10-11-2021",

url = "https://nllpw.org/workshop/",

}

Tziafas, G , De Saint-Phalle, E , de Vries, W, Egger, C & Caselli, T 2021, A Multilingual Approach to Identify and Classify Exceptional Measures against COVID-19. in N Aletras, I Androutsopoulos, L Barrett, C Goanta & D Preotiuc-Pietro (uitgave), Proceedings of the Natural Legal Language Processing Workshop 2021. Association for Computational Linguistics (ACL), blz. 46-62, Natural Legal Language Processing Workshop 2021, 10/11/2021.

A Multilingual Approach to Identify and Classify Exceptional Measures against COVID-19. / Tziafas, Georgios ; De Saint-Phalle, Eugenie ; de Vries, Wietse et al.
Proceedings of the Natural Legal Language Processing Workshop 2021. uitgave / Nikolaos Aletras; Ion Androutsopoulos; Leslie Barrett; Catalina Goanta; Daniel Preotiuc-Pietro. Association for Computational Linguistics (ACL), 2021. blz. 46-62.

Onderzoeksoutput › Academic › peer review

TY - GEN

T1 - A Multilingual Approach to Identify and Classify Exceptional Measures against COVID-19

AU - Tziafas, Georgios

AU - De Saint-Phalle, Eugenie

AU - de Vries, Wietse

AU - Egger, Clara

AU - Caselli, Tommaso

PY - 2021

Y1 - 2021

N2 - The COVID-19 pandemic has witnessed the implementations of exceptional measures by governments across the world to counteract its impact. This work presents the initial results of an on-going project, EXCEPTIUS, aiming to automatically identify, classify and com- pare exceptional measures against COVID-19 across 32 countries in Europe. To this goal, we created a corpus of legal documents with sentence-level annotations of eight different classes of exceptional measures that are im- plemented across these countries. We evalu- ated multiple multi-label classifiers on a manu- ally annotated corpus at sentence level. The XLM-RoBERTa model achieves highest per- formance on this multilingual multi-label clas- sification task, with a macro-average F1 score of 59.8%.

AB - The COVID-19 pandemic has witnessed the implementations of exceptional measures by governments across the world to counteract its impact. This work presents the initial results of an on-going project, EXCEPTIUS, aiming to automatically identify, classify and com- pare exceptional measures against COVID-19 across 32 countries in Europe. To this goal, we created a corpus of legal documents with sentence-level annotations of eight different classes of exceptional measures that are im- plemented across these countries. We evalu- ated multiple multi-label classifiers on a manu- ally annotated corpus at sentence level. The XLM-RoBERTa model achieves highest per- formance on this multilingual multi-label clas- sification task, with a macro-average F1 score of 59.8%.

M3 - Conference contribution

SP - 46

EP - 62

BT - Proceedings of the Natural Legal Language Processing Workshop 2021

A2 - Aletras, Nikolaos

A2 - Androutsopoulos, Ion

A2 - Barrett, Leslie

A2 - Goanta, Catalina

A2 - Preotiuc-Pietro, Daniel

PB - Association for Computational Linguistics (ACL)

T2 - Natural Legal Language Processing Workshop 2021

Y2 - 10 November 2021

ER -

A Multilingual Approach to Identify and Classify Exceptional Measures against COVID-19

Samenvatting

Conference

Toegang tot document

Handle.net

Vingerafdruk

Datasets

EXCEPTIUS Corpus

Projecten

EXCEPTIUS Exceptional powers in time of Sars-CoV-2 crisis

Citeer dit