We study annotation projection in text classification problems where source documents are published in multiple languages and may not be an exact translation of one another. In particular, we focus on the detection of unfair clauses in privacy policies and terms of service. We present the first English-German parallel asymmetric corpus for the task at hand. We study and compare several language-agnostic sentence-level projection methods. Our results indicate that a combination of word embeddings and dynamic time warping performs best.

Galassi Andrea, D.K. (2020). Cross-lingual Annotation Projection in Legal Texts. International Committee on Computational Linguistics [10.18653/v1/2020.coling-main.79].

Cross-lingual Annotation Projection in Legal Texts

Galassi Andrea
;
Torroni Paolo
2020

Abstract

We study annotation projection in text classification problems where source documents are published in multiple languages and may not be an exact translation of one another. In particular, we focus on the detection of unfair clauses in privacy policies and terms of service. We present the first English-German parallel asymmetric corpus for the task at hand. We study and compare several language-agnostic sentence-level projection methods. Our results indicate that a combination of word embeddings and dynamic time warping performs best.
2020
Proceedings of the 28th International Conference on Computational Linguistics
915
926
Galassi Andrea, D.K. (2020). Cross-lingual Annotation Projection in Legal Texts. International Committee on Computational Linguistics [10.18653/v1/2020.coling-main.79].
Galassi Andrea, Drazewski Kasper, Lippi Marco , Torroni Paolo
File in questo prodotto:
File Dimensione Formato  
Cross-lingual-Annotation-Projection-in-Legal-Texts.pdf

accesso aperto

Descrizione: Published paper
Tipo: Versione (PDF) editoriale
Licenza: Licenza per Accesso Aperto. Creative Commons Attribuzione (CCBY)
Dimensione 405.85 kB
Formato Adobe PDF
405.85 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/782746
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 10
  • ???jsp.display-item.citation.isi??? ND
social impact