Syllable-level neural language model for agglutinative language

Seunghak Yu, Nilesh Kulkarni, Haejun Lee, Jihie Kim

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

10 Scopus citations

Abstract

Language models for agglutinative languages have always been hindered in past due to myriad of agglutinations possible to any given word through various affixes. We propose a method to diminish the problem of out-of-vocabulary words by introducing an embedding derived from syllables and morphemes which leverages the agglutinative property. Our model outperforms character-level embedding in perplexity by 16.87 with 9.50M parameters. Proposed method achieves state of the art performance over existing input prediction methods in terms of Key Stroke Saving and has been commercialized.

Original languageEnglish
Title of host publicationEMNLP 2017 - 1st Workshop on Subword and Character Level Models in NLP, SCLeM 2017 - Proceedings of the Workshop
EditorsManaal Faruqui, Hinrich Schutze, Isabel Trancoso, Yaghoobzadeh Yadollah
PublisherAssociation for Computational Linguistics (ACL)
Pages92-96
Number of pages5
ISBN (Electronic)9781945626913
StatePublished - 2017
EventEMNLP 2017 1st Workshop on Subword and Character Level Models in NLP, SCLeM 2017 - Copenhagen, Denmark
Duration: 7 Sep 2017 → …

Publication series

NameEMNLP 2017 - 1st Workshop on Subword and Character Level Models in NLP, SCLeM 2017 - Proceedings of the Workshop

Conference

ConferenceEMNLP 2017 1st Workshop on Subword and Character Level Models in NLP, SCLeM 2017
Country/TerritoryDenmark
CityCopenhagen
Period7/09/17 → …

Fingerprint

Dive into the research topics of 'Syllable-level neural language model for agglutinative language'. Together they form a unique fingerprint.

Cite this