Hideki Isozaki
I will move to a university soon. Thank you for your cooperation.
Japanese page
I received B.E. (1983), M.E (1986), and Ph.D. (1998)
from the University of Tokyo.
In 1986, I joined NTT. I was a visiting scholar at Robotics
Laboratory of Stanford University in 1990-1991.
I am a senior research scientist, supervisor of NTT Communication Science Laboratories.
Affiliation
Linguistic Intelligence Research Group (new group!)
Innovative Communication Laboratory
NTT
Communication Science Laboratories
2-4, Hikaridai, Seika-cho, Soraku-gun, Kyoto, Japan, 619-0237
Research interests
Now, I am working on the improvement of English-to-Japanese and Japanese-to-English statistical machine translation.
- Our paper on automatic evaluation of machine translation was
accepted by EMNLP-2010.
Beamer PDF
- I presented my paper titled
``
Head Finalization: A Simple Reordering Rule for SOV Languages''
at WMT-2010.
An example of Head Finalization (Flash animation).
- Statistical Natural Language Processing
I am interested in Information Extraction such as Named Entity Recognition and Anaphora Resolution.
SVM shows good performance in different Natural Language Processing tasks.
However, SVM is much slower than other methods.
Therefore, I developed an efficient algorithm for the SVM-based Japanese Named Entity Recognizer.
I also worked on resolution of zero pronouns, or omissions of subjects or objects in Japanese.
- Hideki Isozaki, Katsuhito Sudoh, Hajime Tsukada, Kevin Duh:
Head Finalization: A Simple Reordering Rule for SOV Languages,
Proceedings of the ACL-2010 Joint Fifth Workshop on Statistical Machine Translation and Metrics MATR,
pp.244--251, 2010.
- Hideki Isozaki:
Japanese Named Entity Recognition based on a Simple Rule Generator and Decision Tree Learning, Proceedings of ACL-2001, pp.306--313, 2001.
- Hideki Isozaki, Hideto Kazawa: Efficient Support Vector Classifiers for Named Entity Recognition , Proceedings of COLING-2002, pp.390--396, 2002.
- Hideki Isozaki, Hideto Kazawa, Tsutomu Hirao: A Deterministic Word Dependency Analyzer Enhanced With Preference Learning, COLING-2004, pp.275--281.
- Hideki Isozaki, Tsutomu Hirao: Japanese Zero Pronoun Resolution based on Ranking Rules and Machine Learning, Proceedings of EMNLP-2003, pp.184--191.
- Jun Suzuki, Hideki Isozaki: Sequence and Tree Kernels with Statistical Feature Mining,
NIPS-2005.
- Question Answering System
We built a Question Answering system NAZEQA that answers `why' questions.
We also built a question answering system SAIQA, System for Advanced Interactive Question Answering.
We proposed a document retrieval method DIDF for question answering.
We also built a Japanese-English Cross-Lingual Question Answering system.
- Ryuichiro Higashinaka and Hideki Isozaki:
NTT's Question Answering System for NTCIR-6 QAC-4,
Proc. of the 6th NTCIR Workshop Meeting on Evaluation of Information
Access Technologies (NTCIR-6), pp.460--463, 2007.
-
- Hideki Isozaki:
An Analysis of a High Performance Japanese Question Answering System,
ACM Transaction on Asian Language Information Processing, Vol.4, No.3, pp.263--279, 2005.
- Hideki Isozaki, Katsuhito Sudoh, Hajime Tsukada:
NTT's Japanese-English Cross-Lingual Question Answering System, Working Notes of the Fifth NTCIR Workshop Meeting, pp.186--193
- BDD
Our paper ``On the Properties of Combination Set Operations''
(with Prof. Hiroshi G. Okuno (Kyoto Univ.) and
Prof Shinichi MInato (Hokkaido Univ.))
was cited in Prof. Donald E. Knuth's
``The Art of Computer Programming'' volume 4, fascicle 1,
and now, it is called ``family algebra.''
- 湊先生のもうひとつのBDDツールBem-IIは、
要約の研究で利用させてもらいました。
- Nested Belief
I proposed fast algorithms for computing nested belief from observation.
- Hideki Isozaki and Yoav Shoham: A mechanism for reasoning about time and belief, Proceedings of the International Conference on Fifth Generation Computer Systems (FGCS), pp.694--701, Ohmsha, 1992.
- Hideki Isozaki: Reasoning About Belief based on Common Knowledge of Observability of Actions, Proceedings of the First International Conference on Multi-Agent Systems (ICMAS), pp.193--200, 1995.
- Hideki Isozaki and Hirofumi Katsuno: A Semantic Characterization of an Algorithm for Estimating Others' Beliefs from Observation, Proceedings of the National Conference on Artificial Intelligence (AAAI), pp.543--549, 1996.
- Hideki Isozaki and Hirofumi Katsuno: Observability-based Nested Belief Computation for Multiagent Systems and Its Formalization, Intelligent Agent IV, Springer LNAI, 1757, pp.24--41, 2000.
Activity
Awards
- IPSJ Yamashita SIG Research Award (2003)
Isozaki, Kazawa; ``Speeding up Named Entity Recognition based on Support Vector Machines''
- IPSJ Best Paper Award (2004)
Isozaki, Kazawa: ``Speeding up Support Vector Machines for Named Entity
Recognition''
- AFNLP Meritorious Asian NLP Paper Award (2006),
Sudoh, Tsukada, Isozaki: ``Incorporating Speech Recognition Confidence into Discriminative
Named Entity Recognition of Speech Data'', (COLING/ACL 2006 paper).
- NLP-2006 Best Presentation Award (2006)
Suzuki, Isozaki: ``Training Conditional Random Fields based on Minimum Classification Error''
- NLP-2007 Best Presentation Award (2007)
Suzuki, Fujino, Isozaki: ``Semi-Supervised Structured Output Sequence Learning based on
data distribution''
(C) Hideki Isozaki 1996-2011, All rights reserved.