Geographical Distance Is The New Hyperparameter: A Case Study Of Finding The Optimal Pre-trained Language For English-isiZulu Machine Translation.

Muhammad Umair Nasir, Innocent Mchechesi


Abstract
Stemming from the limited availability of datasets and textual resources for low-resource languages such as isiZulu, there is a significant need to be able to harness knowledge from pre-trained models to improve low resource machine translation. Moreover, a lack of techniques to handle the complexities of morphologically rich languages has compounded the unequal development of translation models, with many widely spoken African languages being left behind. This study explores the potential benefits of transfer learning in an English-isiZulu translation framework. The results indicate the value of transfer learning from closely related languages to enhance the performance of low-resource translation models, thus providing a key strategy for low-resource translation going forward. We gathered results from 8 different language corpora, including one multi-lingual corpus, and saw that isiXhosa-isiZulu outperformed all languages, with a BLEU score of 8.56 on the test set which was better from the multi-lingual corpora pre-trained model by 2.73. We also derived a new coefficient, Nasir’s Geographical Distance Coefficient (NGDC) which provides an easy selection of languages for the pre-trained models. NGDC also indicated that isiXhosa should be selected as the language for the pre-trained model.
Anthology ID:
2022.mia-1.1
Volume:
Proceedings of the Workshop on Multilingual Information Access (MIA)
Month:
July
Year:
2022
Address:
Seattle, USA
Editors:
Akari Asai, Eunsol Choi, Jonathan H. Clark, Junjie Hu, Chia-Hsuan Lee, Jungo Kasai, Shayne Longpre, Ikuya Yamada, Rui Zhang
Venue:
MIA
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1–8
Language:
URL:
https://aclanthology.org/2022.mia-1.1
DOI:
10.18653/v1/2022.mia-1.1
Bibkey:
Cite (ACL):
Muhammad Umair Nasir and Innocent Mchechesi. 2022. Geographical Distance Is The New Hyperparameter: A Case Study Of Finding The Optimal Pre-trained Language For English-isiZulu Machine Translation.. In Proceedings of the Workshop on Multilingual Information Access (MIA), pages 1–8, Seattle, USA. Association for Computational Linguistics.
Cite (Informal):
Geographical Distance Is The New Hyperparameter: A Case Study Of Finding The Optimal Pre-trained Language For English-isiZulu Machine Translation. (Nasir & Mchechesi, MIA 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-4/2022.mia-1.1.pdf