Mamoru Komachi

Computational Linguistics Laboratory
Graduate School of Information Science
Nara Institute of Science and Technology
8916-5 Takayama, Ikoma, Nara 630-0192, Japan

komachi--at--is.naist.jp

Research Interests

• Large-scale natural language processing using the Web
• Statistical natural language processing

Current projects include large-scale natural language processing using the web, especially in the field of semantics analysis and information extraction. I am working on large corpora to extract semantic knowledge with minimal human annotation cost.

I am also interested in statistical machine translation (SMT). We are working on incorporating syntactic and semantic information to SMT framework.

Education

Ph.D., March 2010 (Best Student Award)

Graduate School of Information Science, Nara Institute of Science and Technology

Dissertation title: Graph-Theoretic Approaches to Minimally-Supervised Natural Language Learning

Supervisor: Professor, Yuji Matsumoto

M.E., March 2007 (Best Student Award)

Graduate School of Information Science, Nara Institute of Science and Technology

Thesis title: Argument Structure Analysis of Event Nouns Based on Noun-verb Co-occurences and Noun Phrase Patterns

Supervisor: Professor, Yuji Matsumoto

B.A., March 2005

College of Arts and Sciences, University of Tokyo, Komaba

Thesis title: The Language Policy in Colonial Taiwan (in Japanese)

Thesis advisor: Associate Professor, Takuji Okamoto

Professional Experience

April 2010-present

Assistant Professor

Nara Institute of Science and Technology, Japan

I am working on various kinds of natural language processing tasks and data mining tasks.

May 2010-September 2011

Software Engineer

Apple Japan, Inc.

I workedas a rsearch consultant for natural language processing.

(Professional experience prior to 2010)

Journal Papers

• Ryu Iida (Tokyo Institute of Technology), Mamoru Komachi, Naoya Inoue, Kentaro Inui and Yuji Matsumoto. Annotating Predicate-Argument Relations and Anaphoric Relations: Findings from the Building of the NAIST Text Corpus. Journal of Natural Language Processing (In Japanese), Vol.17, No.2, pp.25-50, April 2010.
• Mamoru Komachi, Ryu Iida (Tokyo Institute of Technology), Kentaro Inui and Yuji Matsumoto. Argument Structure Analysis of Event-nouns Using Lexico-syntactic Patterns of Noun Phrases. Journal of Natural Language Processing (In Japanese), Vol.17, No.1, pp.141-159, January 2010.
• Mamoru Komachi, Taku Kudo (Google), Masashi Shimbo and Yuji Matsumoto. Semantic Drift in Espresso-style Bootstrapping: Graph-theoretic Analysis and Evaluation in Word Sense Disambiguation. Transactions of the Janapese Society of Artificial Intelligence (In Japanese), Vol.25, No.2, pp.233-242, January 2010.
• Mamoru Komachi, Shimpei Makimoto (Yahoo Japan), Kei Uchiumi (Yahoo Japan), Manabu Sassano (Yahoo Japan). Learning Semantic Categories from Search Clickthrough Logs Using Laplacian Label Propagation. Transactions of the Japanese Society of Artificial Intelligence (in Japanese), Vol.25, No.1, pp.196-205, January 2010.
• Mamoru Komachi and Hisami Suzuki (Microsoft Research). Improving Semi-Supervised Acquisition of Semantic Knowledge from Query Logs. Transactions of the Japanese Society of Artificial Intelligence (in Japanese), Vol.23, No.3, pp.217-225, March 2008.
• Mamoru Komachi. The Language Policy in Colonial Taiwan: Ando Masatsugu's Monolingualism and Bilingualism The Japanese Journal for the History of Science and Technology (in Japanese), No.8, pp.31-55, 2005.

Conference Papers (Refereed)

• Toshikazu Tajiri, Mamoru Komachi and Yuji Matsumoto. Tense and Aspect Error Correction for ESL Learners Using Global Context. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics: Short papers (oral). Jeju, Korea, July 2012. (to appear)
• Toshinobu Ogiso (National Institute for Japanese Language and Linguistics), Mamoru Komachi, Yasuharu Den (Chiba University) and Yuji Matsumoto. UniDic for Early Middle Japanese: an Electronic Dictionary for Morphological Analysis of Classical Japanese. In Proceedings of the 8th International Conference on Language Resources and Evaluation (LREC): Posters. Istanbul, Turkey, May 2012. (to appear)
• Seiji Kasahara, Mamoru Komachi, Masaaki Nagata (NTT) and Yuji Matsumoto. Error Correcting Romaji-Kana Conversion for Japanese Language Education. In Proceedings of the Workshop on Text Input Methods (WTIM 2011): Short papers (oral), pp.38-42. Chaing Mai, Thailand, November 2011.
• Tomoya Mizumoto, Mamoru Komachi, Masaaki Nagata (NTT) and Yuji Matsumoto. Mining Revision Log of Language Learning SNS for Automated Japanese Error Correction of Second Language Learners. In Proceedings of the 5th International Joint Conference on Natural Language Processing (oral), pp.147-155. Chaing Mai, Thailand, November 2011.
• Yuta Hayashibe, Mamoru Komachi and Yuji Matsumoto. Japanese Predicate Argument Structure Analysis Exploiting Argument Position and Type. In Proceedings of the 5th International Joint Conference on Natural Language Processing (oral), pp.201-209. Chaing Mai, Thailand, November 2011.
• Teruaki Oka, Mamoru Komachi, Toshinobu Ogiso (National Institute for Japanese Language and Linguistics) and Yuji Matsumoto. Automatic Labeling of Voiced Consonants for Morphological Analysis of Modern Japanese Literature. In Proceedings of the 5th International Joint Conference on Natural Language Processing (oral), pp.292-300. Chaing Mai, Thailand, November 2011.
• Kei Uchiumi (Yahoo Japan), Mamoru Komachi, Keigo Machinaga (Yahoo Japan), Toshiyuki Maezawa (Yahoo Japan), Toshinori Satou (Yahoo Japan) and Yoshinori Kobayashi (Google). Japanese Abbreviation Expansion with Query and Clickthrough Logs. In Proceedings of the 5th International Joint Conference on Natural Language Processing (oral), pp.410-419. Chaing Mai, Thailand, November 2011.
• Kohei Ozaki, Masashi Shimbo, Mamoru Komachi and Yuji Matsumoto. Using the Mutual k-Nearest Neighbor Graphs for Semi-supervised Classification on Natural Language Data. In Proceedings of the Fifteenth Conference on Computational Natural Language Learning (CoNLL-2011), pp.154-162. Portland, USA, June 2011.
• Tetsuo Kiso, Masashi Shimbo, Mamoru Komachi and Yuji Matsumoto. HITS-based Seed Selection and Stop List Construction for Bootstrapping. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies (ACL HLT 2011): Short Paper, pp.30-36. Portland, USA, June 2011.
• Mamoru Komachi, Shimpei Makimoto (Yahoo Japan), Kei Uchiumi (Yahoo Japan), Manabu Sassano (Yahoo Japan). Learning Semantic Categories from Clickthrough Logs. In Proceedings of the Joint conference of the 47th Annual Meeting of the Association for Computational Linguistics and the 4th International Joint Conference on Natural Language Processing of the Asian Federation of Natural Language Processing (ACL-IJCNLP 2009): Short Papers, pp.189-192. Singapore, August 2009.
• Mamoru Komachi, Taku Kudo (Google), Masashi Shimbo and Yuji Matsumoto. Graph-based Analysis of Semantic Drift in Espresso-like Bootstrapping Algorithms. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2008), pp.1011-1020. Honolulu, USA, October 2008.
• Masaki Noguchi (Tokyo Institute of Technology), Kenta Miyoshi (Tokyo Institute of Technology), Takenobu Tokunaga (Tokyo Institute of Technology), Ryu Iida, Mamoru Komachi and Kentaro Inui. Multiple Purpose Annotation using SLAT - Segment and Link-based Annotation Tool -. In Proceedings of the 2nd Linguistic Annotation Workshop, pp.61-64, May 2008.
• Mamoru Komachi and Hisami Suzuki (Microsoft Research). Minimally Supervised Learning of Semantic Knowledge from Query Logs. Proceedings of the 3rd International Joint Conference on Natural Language Processing (IJCNLP-08), pp.358-365. Hyderabad, India, January 2008.
• Mamoru Komachi, Ryu Iida, Kentaro Inui and Yuji Matsumoto. Learning Based Argument Structure Analysis of Event-nouns in Japanese. Proceedings of the Conference of the Pacific Association for Computational Linguistics (PACLING), pp.120-128. Melbourne, Australia, September 2007.
• Ryu Iida, Mamoru Komachi, Kentaro Inui, Yuji Matsumoto. Annotating a Japanese Text Corpus with Predicate-Argument and Coreference Relations. Proceedings of the Linguistic Annotation Workshop, pp.132-139. Prague, Czech, June 2007.

Conference Papers (Others)

• Shuhei Kondo, Mamoru Komachi, Yuji Matsumoto, Katsuhito Sudoh (NTT), Kevin Duh (NTT), and Hajime Tsukada (NTT). Learning of Linear Ordering Problems and its Application to J-E Patent Translation in NTCIR-9 PatentMT. In Proceedings of NTCIR-9 Workshop Meeting, pp.641-645, Tokyo, Japan, December 2011.
• Joseph Irwin, Mamoru Komachi and Yuji Matsumoto. Narrative Schema as World Knowledge for Coreference Resulution. In Proceedings of the Fifteenth Conference on Computational Natural Language Learning (CoNLL-2011 Shared Task), pp.32-38. Portland, USA, June 2011.
• Mamoru Komachi, Masaaki Nagata (NTT) and Yuji Matsumoto. NAIST-NTT System Description for Patent Translation Task at NTCIR-7 In Proceedings of the NTCIR-7 Workshop, pp.435-440. Tokyo, Japan, Decemver 2008.
• Mamoru Komachi, Masaaki Nagata and Yuji Matsumoto. Phrase Reordering for Statisitcal Machine Translation Based on Predicate-Argument Structure. Proceedings of the International Workshop on Spoken Language Translation (IWSLT 2006), pp.77-82. Kyoto, Japan, November 2006.

Grants

PI

April 2012-March 2012 JPY 2.0M

Microsoft Research CORE Project, "Learning L1-aware Error Models from the Web"

April 2011-March 2013 JPY 2.7M

JSPS Grant-in-Aid for Young Scientists (B), "Large-scale Parallel Processing for Global Optimization in Natural Language Processing"

April 2011-March 2014 JPY 2.5M

SCAT Research Grant, "Robust Japanese Predicate Argument Structure Analysis for Large-scale Information Extraction"

April 2008-March 2010 JPY 1.0M

Japan Society for the Promotion of Science (JSPS) Research Fellowship for Young Scientists, "Exploiting Large Corpora for Learning-based Argument Structure Analysis of Noun Phrases"

October 2007-March 2008 JPY 1.25M

Creative and International Competitiveness Project Leader, "Robust Statistical Japanese IME with Unknown Word Processing Using Large Corpus", Nara Institute of Science and Technology.

Co-I

April 2011-March 2014

MEXT Grant-in-Aid for Scientific Research (A), "Joint Approach to Natural Language Processing Using Global Information"

November 2011-September 2013

National Institute for Japanese Language, "Statistical and Machine Learning-based Approaches to History of the Japanese Language Research"

Summer 2010-March 2011

MEXT Grant-in-Aid for Scientific Research in Priority Areas, "Japanese Corpus: Automatic Labeling of the Balanced Corpus of Contemporary Written Japanese (BCCWJ)"

July 2005-March 2006

Tomoya Yoshizawa, Keita Oda, Takashi Asari, Yasushi Hikida, Shinra Aida, Tomoaki Okayama, Hiroyuki Onoda, Motoi Washida and Mamoru Komachi: "User-Oriented Package Management System", Exploratory Software Project (Exploratory Youth Program), Information-Technology Promotion Agency, Japan.

Honors and Awards

May 2011

JSAI Best Paper Award, The Japanese Society for Artificial Intelligence.

August 2010

IPSJ Yamashita SIG Research Award, Information Processing Society of Japan. "Learning Semantic Categories from Web Search Logs Using Label Propagation" (One of the best CS awards in Japan)

March 2009

Best Paper Award, the 14th Annual Meeting of the Association for Natural Language Processing, "Definition of Semantic Similarity using Kernel Methods and Generalization of Bootstrapping"

March 2008

Outstanding Paper Award, the 13th Annual Meeting of the Association for Natural Language Processing, "Annotating Predicate-argument and Coreference Relations in Japanese Written Text: From the Experience of Building the NAIST Text Corpus"

November 2005-March 2007

Scholarship of Graduate School of Information Science, Nara Institute of Science and Technology.

Other Academic Activities

• Conference on Empirical Methods in Natural Language Processing and Natural Language Learning (EMNLP-CoNLL 2012) Program Committee (2012)
• The 13th Conference of the European Chapter of the Association for Computational Linguistics (EACL-2012) Program Committee (2012)
• Internatinal Joint Conference on Natural Language Processing (IJCNLP 2011) Program Committee (2011)
• Workshop on Advances in Text Input Methods (WTIM 2011) Program Committee (2011)
• Conference on Computational Natural Language Learning (CoNLL 2011) Program Committee (2011)
• AAAI Conference on Artificial Intelligence (AAAI-11) Program Committee (2011)
• IEICE Transactions of Information and Systems Special Issue: Natural Language Processing and its Applications Reviewer (2009)
• North American Chapter of the Association for Computational Linguistics - Human Language Technologies Conference (NAACL-HLT) 2009 Program Committee
• Conference of the Pacific Association for Computational Linguistics (PACLING) 2008 Session chair

Social Activities

June 2004-

Fink Project: Developer (Awarded Apple WWDC Student Scholarship in 2006 and 2009)

July 2003-October 2007

Gentoo Project: Developer (CJK team)

Other Information

• Natural languages: Japanese (native), English (advanced: TOEFL CBT 266, TOEFL iBT 98; TOEIC 960), Mandarin Chinese and German (reading)
• Programming languages: Python (4+yrs), C++ (7+yrs), Perl (7+yrs), Bash (7+yrs), C# (4+yrs)
• Platforms (currently developing on): Linux (Ubuntu, Gentoo), Mac OS X