Learning variable length units for SMT between related languages via Byte Pair Encoding

Anoop Kunchukuttan, Pushpak Bhattacharyya


Abstract
We explore the use of segments learnt using Byte Pair Encoding (referred to as BPE units) as basic units for statistical machine translation between related languages and compare it with orthographic syllables, which are currently the best performing basic units for this translation task. BPE identifies the most frequent character sequences as basic units, while orthographic syllables are linguistically motivated pseudo-syllables. We show that BPE units modestly outperform orthographic syllables as units of translation, showing up to 11% increase in BLEU score. While orthographic syllables can be used only for languages whose writing systems use vowel representations, BPE is writing system independent and we show that BPE outperforms other units for non-vowel writing systems too. Our results are supported by extensive experimentation spanning multiple language families and writing systems.
Anthology ID:
W17-4102
Volume:
Proceedings of the First Workshop on Subword and Character Level Models in NLP
Month:
September
Year:
2017
Address:
Copenhagen, Denmark
Editors:
Manaal Faruqui, Hinrich Schuetze, Isabel Trancoso, Yadollah Yaghoobzadeh
Venue:
SCLeM
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
14–24
Language:
URL:
https://aclanthology.org/W17-4102
DOI:
10.18653/v1/W17-4102
Bibkey:
Cite (ACL):
Anoop Kunchukuttan and Pushpak Bhattacharyya. 2017. Learning variable length units for SMT between related languages via Byte Pair Encoding. In Proceedings of the First Workshop on Subword and Character Level Models in NLP, pages 14–24, Copenhagen, Denmark. Association for Computational Linguistics.
Cite (Informal):
Learning variable length units for SMT between related languages via Byte Pair Encoding (Kunchukuttan & Bhattacharyya, SCLeM 2017)
Copy Citation:
PDF:
https://aclanthology.org/W17-4102.pdf