2020 Fiscal Year Final Research Report
Developing an Environment to Suppot the Revision of English Translations of Japanese Statutes due to Amendments: A Method based on Hybrid Machine Translation
| Project/Area Number |
18H03492
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 90020:Library and information science, humanistic and social informatics-related
|
|---|
| Research Institution | Nagoya University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2018-04-01 – 2021-03-31
|
| Keywords | 法情報処理 / 法令翻訳 / 法制執務 / 機械翻訳 / 自然言語処理 |
|---|
| Outline of Final Research Achievements |
This research aims to develop a machine translation technology that supports the revision of English translations of Japanese statutes due to their partial amendments, which contributes to immediate and international dissemination of legal information on Japan necessary for the globalization of society. In particular, as well as adequate and fluent translation, we require "focal" translation: only the translation for the expressions that are amended in the statutory sentences should be revised without revising the others.
To establish this, we developed a hybrid machine translation technology that combines statistical machine translation using a translation memory and neural machine translation. By compiling and using a corpus which consists of Japanese-English bilingual statutory sentences for pre- and post-amendment, we showed that our proposed method is effective for supporting the revision of English translations of Japanese statutes.
|
| Free Research Field |
自然言語処理
|
| Academic Significance and Societal Importance of the Research Achievements |
社会のグローバル化に伴い，わが国の法情報の国際的発信が国内外から求められている．しかし，法令の新規制定や改正に対する英訳作業の遅れや訳語の一貫性の不徹底などの課題がある．本研究は，法令の一部改正に伴う英訳修正を支援する機械翻訳手法を開発し，英訳作業の効率化・高精度化に貢献した．また，訳文の妥当性と流暢性だけでなく，訳文の修正極小性も訳文の評価として新たに要求することにより，一過性の翻訳だけでなく，原文の修正に対して動的・継続的に対応する機械翻訳技術の開発という研究課題を新たに設定し，翻訳メモリを用いた統計的機械翻訳とニューラル機械翻訳を融合したハイブリッド機械翻訳という新たな技術を開発した．
|
