Achievements |

Achievements by the UTX team

In 2011, the UTX team (Working Group 3, AAMT MT Research Committee) received the 6th AAMT Nagao Award. The Nagao Award commends individuals or groups who have contributed to research and development leading to or facilitating commercialization of machine translation systems. The reason for the award: "Working Group 3 has developed and released an easy-to-use, open dictionary format UTX. UTX provides effective means for improving machine translation accuracy as well as the quality of computer assisted translation and interoperability of user-created dictionaries. UTX is valuable because it has the potential to grow into an international standard in addition to its use in various specialized domains such as ICT localization, open source efforts, education, government, medical, legal, and many other areas."

External activities of the UTX team

In 2010, the UTX team created English-Japanese and Japanese-English bilingual UTX glossaries in legal and medical domains (containing 5451 words and 27126 words respectively) and made them available for free.

Since 2011, working in cooperation with the translation industry, AAMT and its UTX team had a booth at the annual JTF (Japan Translation Federation) Translation Festival.

Academic papers by UTX team members (at international academic meetings with peer review)

Francis Bond et al. (2009) "Sharing User Dictionaries Across Multiple Systems with UTX-S" (PDF) in Second International Workshop on Intercultural Collaboration (IWIC2009), Stanford.
- (The researchers created a basic conversion script between UTX and other formats and demonstrated that the translation accuracy is improvable with a UTX glossary containing limited numbers of words.)
Seiji Okura et al. (2011) "UTX 1.11, a Simple and Open User Dictionary/Terminology Standard, and its Effectiveness with Multiple MT Systems" (PDF) in MT Summit 2011 (Xiamen).

Other papers, workshop presentations, and magazine articles by UTX team members

Okura, S. et al. (2008). Introduction to UTX, a specification for a shared user dictionary. 14th Annual Meeting of Association for Natural Language Processing (Tokyo).
Okura, S. (2009, November). Presentation at the LISA China Focus (Beijing).
Yamamoto, Y. (2011, February). Presentation at the LISA Open Standards Summit. (Boston).
Okura, S. et al. (2011). 「翻訳支援のためのシンプルでオープンな辞書仕様UTX-Simple 1.10」[Simple and open dictionary specification UTX-Simple 1.10 for computer-assisted translation] (PDF). 17th Annual Meeting of Association for Natural Language Processing (Tokyo)
Yamamoto, Y. (2011, June). How to create glossaries in UTX. MultiLingual.

Yamamoto, Y. (2012). 「実務日本語とUTXから特許明細書の改善を考える」[Improving the writing of patent specifications in view of Practical Japanese and UTX]. Japio Year Book 2012.
Yamamoto, Y. (2013). UTX (universal terminology eXchange): a simple terminology format. GALAxy Newsletter.
Yamamoto, Y. (2013). 「日本語文章の課題と前編集手法―用語集形式UTX と実務日本語の観点から」[Issues of Japanese writing and pre-edit approaches to address these issues– in view of UTX and Practical Japanese]. Japio Year Book 2013.
Yamamoto, Y. (2013). 「シンプルな用語集形式UTXとその活用 ―ビッグデータと『グッドデータ』のいい関係」[UTX, a Simple Glossary Format, and Its Applications: How 'Good Data' Interact with Big Data]. The 5th Symposium on Collective Knowledge.
Yamamoto, Y. (2014). 「用語集形式UTXの現況と実務日本語・百半ルール――実際的な用語集と表記規則を実現するには」[Glossary format UTX and the "Break 100" rule of Practical Japanese]. Japio Year Book 2014.
Yamamoto, Y. (2015, June). UTX, a simple glossary format. Presentation at the ISO/TC37 meeting (Matsue).
Yamamoto, Y. (2015, October). UTX: a simple standardized glossary format. Presentation at the 15th China-Japan Natural Language Processing Collaboration Promotion Conference (Aomori).
Yamamoto, Y. (2015). 「UTXと、用語管理、翻訳品質、および標準化の関連性―ISO/TC37松江国際会議に参加して」[UTX and its relation with terminology management, translation quality, and standardization – Report on the ISO/TC37 2015 annual working meetings and plenary]. Japio Year Book 2015.
Yamamoto, Y. (2016). 「用語データを共有・再利用するための用語集形式の標準規格UTX」[UTX, a Standardized Glossary Format for Sharing and Reusing Glossary Data]. 22th Annual Meeting of Association for Natural Language Processing (Sendai)
Yamamoto, Y. (2016). 「体系的翻訳・文書品質管理での高品質言語資産としてのUTX用語集形式」[UTX glossary format as high-quality language assets in systematic translation and document quality control]. IPSJ SIG Technical Report.
Yamamoto, Y. (2017).「Win-Winの機械翻訳最新活用―ニューラル機械翻訳をUTX用語データでポストエディットする方法」[Machine translation usage for Win-Win relationship - A post edit method to check neural MT results with UTX terminology format] STC東京特別セミナー
Yamamoto, Y. (2017). 「ニューラルMTを構造化用語データUTXで補完して翻訳品質を改善する簡便なポストエディット手法」[A simple post edit method by checking neural MT results with terminology data format UTX to improve translation quality]. 23th Annual Meeting of Association for Natural Language Processing (Tsukuba)

Implementations of UTX

UTX format was adopted as a dictionary format in the Patent Office machine translation research. According to the research report, a Chinese-Japanese bilingual dictionary containing about 800,000 words was compiled, and a huge improvement was observed especially in the translation accuracy of compound nouns. Based on this dictionary and other dictionaries, a Chinese-Japanese bilingual dictionary containing about 1,000,000 entries was compiled in 2014. In 2015, a Chinese-Japanese bilingual dictionary containing about 2.2 million entries was compiled, and released for research purposes at the ALAGIN Language and Speech Resources website.