• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Forest-based neural machine translation for improving translation performance between distant language pairs

Research Project

Project/Area Number 18K18110
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 61030:Intelligent informatics-related
Research InstitutionEhime University

Principal Investigator

Tamura Akihiro  愛媛大学, 理工学研究科(工学系), 助教 (20804165)

Project Period (FY) 2018-04-01 – 2020-03-31
Project Status Completed (Fiscal Year 2019)
Budget Amount *help
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2019: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Fiscal Year 2018: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
Keywordsニューラル機械翻訳 / 構文森 / ニューラルネットワーク / 機械翻訳 / Transformer / RNN / Transformer
Outline of Final Research Achievements

This research aims to improve the performance of neural network-based machine translation (NMT) by using a source-side syntax forest, which is a compact representation of many parse trees. The study has proposed a novel method for incorporating source-side syntax forests into recurrent neural network-based NMT and Transformer-based NMT. The evaluations show that the syntax forests improve the English-Japanese translation performance and the proposed model achieves a state-of-the-art performance.

Academic Significance and Societal Importance of the Research Achievements

近年のグローバル化の進展とともに、外国語の利活用を支援する機械翻訳の需要が高まっている。しかし、機械翻訳では、構造が異なる言語間の翻訳は難しく、その翻訳性能の改善が大きな課題の一つとなっている。本研究では、その課題を解決するため、翻訳元の文の構文森の情報をNMTで活用する初めての試みに取り組んだ。そして、構文森を活用することにより、構造が異なる言語間の代表例である英語と日本語間の翻訳性能を改善できることを示し、今後の機械翻訳の研究開発において、構文森を活用する重要性を示唆した。

Report

(3 results)
  • 2019 Annual Research Report   Final Research Report ( PDF )
  • 2018 Research-status Report
  • Research Products

    (8 results)

All 2020 2019 2018

All Journal Article (3 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 3 results,  Open Access: 3 results) Presentation (5 results) (of which Int'l Joint Research: 5 results)

  • [Journal Article] Syntax-based Transformer for Neural Machine Translation2020

    • Author(s)
      Chunpeng Ma, Akihiro Tamura, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao
    • Journal Title

      Journal of Natural Language Processing

      Volume: 27(2)

    • NAID

      130007904713

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Encoder-Decoder Attention ≠ Word Alignment: Axiomatic Method of Learning Word Alignments for Neural Machine Translation2020

    • Author(s)
      Chunpeng Ma, Akihiro Tamura, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao
    • Journal Title

      Journal of Natural Language Processing

      Volume: 27(3)

    • NAID

      130007956043

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] CKYに基づく畳み込みアテンション構造を用いたニューラル機械翻訳2019

    • Author(s)
      渡邊 大貴, 田村 晃裕, 二宮 崇, Teguh Bharata Adji
    • Journal Title

      自然言語処理

      Volume: 26(1) Pages: 207-230

    • NAID

      130007663695

    • Related Report
      2018 Research-status Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Presentation] Dependency-Based Relative Positional Encoding for Transformer NMT2019

    • Author(s)
      Yutaro Omote, Akihiro Tamura, Takashi Ninomiya
    • Organizer
      International Conference on Recent Advances in Natural Language Processing 2019
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Dependency-Based Self-Attention for Transformer NMT2019

    • Author(s)
      Hiroyuki Deguchi, Akihiro Tamura, Takashi Ninomiya
    • Organizer
      International Conference on Recent Advances in Natural Language Processing 2019
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Improving Neural Machine Translation with Neural Syntactic Distance2019

    • Author(s)
      Chunpeng Ma, Akihiro Tamura, Masao Utiyama, Eiichiro Sumita, Tiejun Zhao
    • Organizer
      2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Multimodal Neural Machine Translation Using CNN and Transformer Encoder2019

    • Author(s)
      Hiroki Takushima, Akihiro Tamura, Takashi Ninomiya, Hideki Nakayama
    • Organizer
      20th International Conference on Computational Linguistics and Intelligent Text Processing
    • Related Report
      2018 Research-status Report
    • Int'l Joint Research
  • [Presentation] Forest-Based Neural Machine Translation2018

    • Author(s)
      Chunpeng Ma, Akihiro Tamura, Masao Utiyama, Tiejun Zhao, Eiichiro Sumita
    • Organizer
      56th Annual Meeting of the Association for Computational Linguistics
    • Related Report
      2018 Research-status Report
    • Int'l Joint Research

URL: 

Published: 2018-04-23   Modified: 2021-02-19  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi