Akira Shimazu

Professor　 Akira Shimazu

He joined JAIST in April, 1997. He received B.S., M.S. and Ph.D. degrees in 1971, 1973 and 1991, respectively, from Kyushu University. From 1973 to 1997, he worked at Musashino Electrical Communication Laboratories of Nippon Telegram and Telephone Public Corporation and at Basic Research Laboratories of Nippon Telegraph and Telephone Corporation.

[English/Japanese]

Specialties

Natural Language Processing, Dialog Processing, Machine Translation, Legal Engineering

Research Interest

Language processing technology which facilitates language communication, both between humans and between a human and a machine, and helps human activities for information acquisition and transfer.

Current Research Topics

Legal texts processing as a research of Legal Engineering: Research of natural language processing techniques for verifying logical consistency of legal texts. The research clarifies methods for extracting LF from texts, analyzing discourse, and paraphrasing texts. The method utilize a viewpoint of software engineering.
Analysis of human language use: An utterance in spoken language is not necessarily represented by a sentence as in written language, and information is conveyed from speaker to hearer in rather small phrasal units with disfluencies such as fillers and self-repairs. Such characteristics are pertinent to human information processing and communication, and analyzed for an advanced natural language processing system.
Dynamic explanation of information: Research which clarifies principles and techniques for explaining information as a human explains information to a recipient in spoken language. The research is based on the analysis of human language use. Not only spoken language systems but also application to web access are within the scope.
Natural language processing based on statistical models: Researches on semantic analysis, machine translation, ambiguity resolution based on corpus and machine learning.

Selected Publications

Akira Shimazu and Hitoshi Iida. Experimental Japanese Language Question Answering System MSSS78. In Proc. of the Sixth International Joint Conference on Artificial Intelligence, pp.803-808, 1979.
Akira Shimazu, Shozo Naito and Hirosato Nomura. Semantic Structure Analysis of Japanese Noun Phrases with Adnominal Particles. In Proc. of the 25th Annual Meeting of the Association for Computational Linguistics, pp.123-130, 1987.
Akira Shimazu. Scheduling of Parsing Processes -From the viewpoint of a Mental OS-. In Proc. of the Second Japan-Australia Joint Symposium on Natural Language Processing, pp. 114-121, 1991.
Kohji Dohsaka and Akira Shimazu. A Computational Model of Incremental Utterance Production in Task-Oriented Dialogues. In Proc. of the 16th International Conference on Computational Linguistics, pp.304-309, 1996.
Masahito Kawamori, Akira Shimazu. Speech Acts and Dialogue Control in Task-oriented Spontaneous Dialogue. In the Proceedings ofAmsterdam Workshop on the Semantics and Pragmatics of Dialogue (Amstelogue'99), 1999.
Mikio Nakano, Akira Shimazu. Parsing Utterances Including Self-Repairs. In Yorick Wilks (ed.), Machine Conversations. Kluwer Academic Publishers, 1999.
M. T. Kone, A. Shimazu and T. Nakajima. The State of the Art in Agent Communication Languages. Knowledge and Information Systems Journal , Vol.2, No.3, pp.259-284, Springer-Verlag, 2000.
H. Murakoshi, A. Shimazu, K. Ochimizu. Construction of Deliberation Structure in Email Communication. Computational Intelligence, Vol.16, No.4, pp.570-577, 2000.
Hideki Uematsu, Akira Shimazu, Manabu Okumura. Generation of 3D CG Animations from Recipe Sentences. In the proceedings of the sixth Natural Language Proceesing of Pacific Rim Symposium, pp.461-466, 2001.
M. L. Nguyen, A. Shimazu, and S. Horiguchi. A New Template Translation Learning Based on Hidden Markov Modeling. WSEAS Transaction on Computers, Issue 1, Volume 3, 256-262, 2004.
Minh Nguyen Le, Susumu Horiguchi, Akira Shimazu, and Bao Tu Ho. Example-based Sentence Reduction Using Hidden Markov Model. ACM Transactions on Asian Language Information Processing. Vol.3, Issue 2, pp.146-158, 2004.
Cuong Anh Le, Akira Shimazu and Van-Nam Huynh. Word Sense Disambiguation by Combining Classifiers with an Adaptive Selection of Context Representation. Journal of Natural Language Processing, Vol.13, No.1, 2006.
Thai Phuong Nguyen, Akira Shimazu. Improving Phrase-Based Statistical Machine Translation with Morpho-Syntactic Analysis and Transformation. In the Proceedings of the 7th Conference of the Association for Machine Translation in the Americas (AMTA 2006), pp.138-147, 2006.
Tri Thanh Nguyen, Minh Le Nguyen, Akira Shimazu. Using Semi-Supervised Learning for Question Classification. In the Proceedings of the 21st International Conference on Computer Processing of Oriental Languages (ICCPOL 2006), pp.31-41, 2006.

Natural Language Processing Laboratories