Your browser does not seem to support some of the display features. As a result, some content may be hidden nor you may not be able to perform some actions.

Your browser does not seem to support JavaScript. As a result, some content may be hidden nor you may not be able to perform some actions.

picture
Junta Mizuno Ph.D (Engineering)

Senior Researcher
Data-driven Intelligent System Research Center,
Universal Communication Research Institute, NICT
Resilient ICT Research Center Applications Laboratory
junta-m at_mark_here nict.go.jp

Biography

Apr. 2018 - Current Senior Researcher, NICT
Apr. 2013 - Mar. 2018 Researcher, NICT
Apr. 2012 - Mar. 2013 Researcher, Tohoku University
Apr. 2010 - Mar. 2012 Special Research Student in Graduate School of Information Science, Tohoku University
Apr. 2009 - Mar. 2012 Ph.D. Student in Graduate School of Information Science, Nara Institute of Science and Technology
Apr. 2007 - Mar. 2009 Master Student in Graduate School of Information Science, Nara Institute of Science and Technology
Apr. 2006 - Mar. 2007 Research Student in Information and Communication Science, Toyohashi University of Technology
Apr. 2002 - Mar. 2006 Bachelor Student in Information and Communication Science, Toyohashi University of Technology

Research Topics

Publications

Recent Publications

WISDOM X, DISAANA and D-SUMM: Large-scale NLP Systems for Analyzing Textual Big Data
Junta Mizuno, Masahiro Tanaka, Kiyonori Ohtake, Jong-Hoon Oh, Julien Kloetzer, Chikara Hashimoto and Kentaro Torisawa
Proceedings of the 26th International Conference on Computational Linguistics (COLING 2016) (Demo Track), Osaka, Japan, December, 2016. [PDF (ACL)]
Stance Classification by Recognizing Related Events about Targets
Akira Sasaki, Junta Mizuno, Naoaki Okazaki and Kentaro Inui
Proceedings of the IEEE/WIC/ACM International Conference on Web Intelligence, pp. 582-587, October 2016.
Improving Question Answering of Disaster-Information Analyzer (DISAANA) Using Modality Analysis
Junta Mizuno, Jun Goto, Kiyonori Ohtake, Takuya Kawada, Kentaro Torisawa, Kloetzer Julien, Masahiro Tanaka, Chikara Hashimoto, and Akitoshi Okumura
IPSJ Journal: CDS18, Vol.6, No.1, pp.106-120, May 2016. (in Japanese)
Understanding the technical issues in Japanese factuality analysis through error analysis
Kazuya Narita, Junta Mizuno, Yudai Kamioka, Miwa Kanno and Kentaro Inui
Journal of Natural Language Processing. Vol.22, No.5, pp.392-432, December 2015. (in Japanese)
Recognizing Complex Negation on Twitter
Junta Mizuno, Canasai Kruengkrai, Kiyonori Ohtake, Chikara Hashimoto, Kentaro Torisawa, Julien Kloetzer and Kentaro Inui
Proceedings of The 29th Pacific Asia Conference on Language, Information and Computation (PACLIC-29), pp.544-552, October 2015. [PDF]
Full Paper List

Awards

IPSJ Yamashita SIG Research Award
Maejima Hisoka Award 2016
CDS14 Best Presentation Award
The 14th (2015) Docomo Mobile Science Award
Twitter #Data Grants
AMT2014 Best Paper Award
Mining False Information on Twitter for a Major Disaster Situation
Keita Nabeshima, Junta Mizuno, Naoaki Okazaki and Kentaro Inui
In proceedings of The 2014 International Conference on Active Media Technology (AMT2014), pp.96-109, August, 2014
The Best Paper Award of the Journal of Natural Language Processing 2013
Extracting False Information on Twitter and Analyzing its Diffusion Processes by using Linguistic Patterns for Correction
Keita Nabeshima, Kento Watanabe, Junta Mizuno, Naoaki Okazaki and Kentaro Inui
Journal of Natural Language Processing, Vol.20, No.3, pp.461-484, 2013
The Best Paper Award of the 16th Annual Meeting of the Association for Natural Language Processing
Statement Map Generation Based on Semantic Relation Recognition
Koji Murakami, Junta Mizuno, Hayato Goto, Megumi Ohki, Suguru Matsuyoshi, Kentaro Inui, Yuji Matsumoyo
16th Annual Meeting of the Association for Natural Language Processing, pp.559-562, March, 2010

Dissertations

Organizing Information Based on Semantic Relation Recognition [PDF]
Doctoral Dissertation, 2012
Nara Institute Science and Technology
Human-Computer Interaction System Using Web News [PDF (in Japanese)]
Master's Thesis, 2009
Nara Institute Science and Technology

Software

Zunda: Japanese Extended Modality Analyzer
文中のイベント(動詞や形容詞など)に対して、その真偽判断(イベントが起こったかどうか)、仮想性(仮定の話かどうか)などを解析することのできる「拡張モダリティ解析器」です。例えば「次郎は大阪に行ったが、太郎は東京には行かず地元に残ろうとした」という文を解析すると、「次郎が大阪に行く」ことは実際に起こったこと(真偽判断:成立)と解析され、「太郎が東京に行く」ことは起こっていないこと(真偽判断:不成立)と解析されます。
GitHub
Dec. 25 / 2013
Written in C++ with Boost

Projects

東日本大震災ビッグデータワークショップ Project 311

2012年9月12日~10月28日にかけて開催された東日本大震災ビッグデータワークショップに参加し,提供データの一つである震災時のツイートデータを高速に検索できる 全文検索エンジンの構築,ツイート中の画像情報に着目して現地被災者を抽出する支援技術の開発( 論文),NICT ユニバーサルコミュニケーション研究所の開発した質問応答システム「一休」の回答に対して,その信憑性を判断する支援技術(言論マップ)の開発などに取り組みました.

研究助成

学会活動

技術研修


授業 (東北大学所属時)