Morphological analysis for unsegmented languages using recurrent neural network language model

Hajime Morita, Daisuke Kawahara, Sadao Kurohashi

研究成果: Conference contribution

46 被引用数 (Scopus)

抄録

We present a new morphological analysis model that considers semantic plausibility of word sequences by using a recurrent neural network language model (RNNLM). In unsegmented languages, since language models are learned from automatically segmented texts and inevitably contain errors, it is not apparent that conventional language models contribute to morphological analysis. To solve this problem, we do not use language models based on raw word sequences but use a semantically generalized language model, RNNLM, in morphological analysis. In our experiments on two Japanese corpora, our proposed model significantly outperformed baseline models. This result indicates the effectiveness of RNNLM in morphological analysis.

本文言語English
ホスト出版物のタイトルConference Proceedings - EMNLP 2015
ホスト出版物のサブタイトルConference on Empirical Methods in Natural Language Processing
出版社Association for Computational Linguistics (ACL)
ページ2292-2297
ページ数6
ISBN(電子版)9781941643327
DOI
出版ステータスPublished - 2015
外部発表はい
イベントConference on Empirical Methods in Natural Language Processing, EMNLP 2015 - Lisbon, Portugal
継続期間: 2015 9月 172015 9月 21

出版物シリーズ

名前Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing

Conference

ConferenceConference on Empirical Methods in Natural Language Processing, EMNLP 2015
国/地域Portugal
CityLisbon
Period15/9/1715/9/21

ASJC Scopus subject areas

  • 計算理論と計算数学
  • コンピュータ サイエンスの応用
  • 情報システム

フィンガープリント

「Morphological analysis for unsegmented languages using recurrent neural network language model」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル