A Syllable-based Technique for Word Embeddings of Korean Words

Sanghyuk Choi, Taeuk Kim, Jinseok Seol, Sang-goo Lee


Abstract
Word embedding has become a fundamental component to many NLP tasks such as named entity recognition and machine translation. However, popular models that learn such embeddings are unaware of the morphology of words, so it is not directly applicable to highly agglutinative languages such as Korean. We propose a syllable-based learning model for Korean using a convolutional neural network, in which word representation is composed of trained syllable vectors. Our model successfully produces morphologically meaningful representation of Korean words compared to the original Skip-gram embeddings. The results also show that it is quite robust to the Out-of-Vocabulary problem.
Anthology ID:
W17-4105
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:
36–40
Language:
URL:
https://aclanthology.org/W17-4105
DOI:
10.18653/v1/W17-4105
Bibkey:
Cite (ACL):
Sanghyuk Choi, Taeuk Kim, Jinseok Seol, and Sang-goo Lee. 2017. A Syllable-based Technique for Word Embeddings of Korean Words. In Proceedings of the First Workshop on Subword and Character Level Models in NLP, pages 36–40, Copenhagen, Denmark. Association for Computational Linguistics.
Cite (Informal):
A Syllable-based Technique for Word Embeddings of Korean Words (Choi et al., SCLeM 2017)
Copy Citation:
PDF:
https://aclanthology.org/W17-4105.pdf