Caveats of Measuring Semantic Change of Cognates and Borrowings using Multilingual Word Embeddings

Clémentine Fourrier, Syrielle Montariol


Abstract
Cognates and borrowings carry different aspects of etymological evolution. In this work, we study semantic change of such items using multilingual word embeddings, both static and contextualised. We underline caveats identified while building and evaluating these embeddings. We release both said embeddings and a newly-built historical words lexicon, containing typed relations between words of varied Romance languages.
Anthology ID:
2022.lchange-1.10
Volume:
Proceedings of the 3rd Workshop on Computational Approaches to Historical Language Change
Month:
May
Year:
2022
Address:
Dublin, Ireland
Editors:
Nina Tahmasebi, Syrielle Montariol, Andrey Kutuzov, Simon Hengchen, Haim Dubossarsky, Lars Borin
Venue:
LChange
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
97–112
Language:
URL:
https://aclanthology.org/2022.lchange-1.10
DOI:
10.18653/v1/2022.lchange-1.10
Bibkey:
Cite (ACL):
Clémentine Fourrier and Syrielle Montariol. 2022. Caveats of Measuring Semantic Change of Cognates and Borrowings using Multilingual Word Embeddings. In Proceedings of the 3rd Workshop on Computational Approaches to Historical Language Change, pages 97–112, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Caveats of Measuring Semantic Change of Cognates and Borrowings using Multilingual Word Embeddings (Fourrier & Montariol, LChange 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.lchange-1.10.pdf
Video:
 https://aclanthology.org/2022.lchange-1.10.mp4
Code
 clefourrier/historical-semantic-change