Summary of Research Project Results Under the JSPS FY2000
"Research for the future Program"



1.Research Institution University of Tokyo
 
2.Research Area Physical and Engineering Science
 
3.Research Field Intelligent Information and Advanced Information Processing
 
4.Term of Project FY1996〜FY2000
 
5.Project Number 96P00502
 
6.Title of Project Natural Language Processing and Natural Language Understanding

7.Projetct Leader
Name Institution,Department Title of Position
Jun-ichi Tsujii University of Tokyo, Graduate School of Science Professor

8.Core Members

Names Institution,Department Title of Position
Kazuhiko Ohe University of Tokyo, Graduate School of Medicine Professor
Sadao Kurahashi Kyoto University, Graduate School of Informatics Lecturer
Takenobu Tokunaga Tokyo Institute of Technology, Graduate School of Information Science and Engineering Associate Professor

9.Cooperating Researchers

Names Institution,Department Title of Position
Kiyoaki Shirai Tokyo Institute of Technology, Graduate School of Information Science and Engineering Research Associate
Yuka Tateishi University of Tokyo, Graduate School of Science Post Doctor
Hideki Mima University of Tokyo, Graduate School of Science Post Doctor

10.Summary of Research Results

The objectives of the project are to establish NLP basic techniques for intelligent information navigation in the network era and to show the feasibility of the current NLP technology. In order to realize the objectives, the following studies have been carried out.
(1) Parsing Technology: By close cooperation with research institutions abroad such as Stanford University and DFKI, an efficient parsing method based on compilation of grammar has been successfully developed. The method shows the best performance in terms of processing time for unification-based grammar formalisms such HPSG. The results are published in a special issue on parsing technology in Journal of Natural Language Engineering. The special issue was coordinated by our group, together with those of Stanford Univ. and DFKI.
(2) English and Japanese grammar: A special method of translation grammar in a formalism to grammar in another formalism has been developed and applied to translation of LTAG grammar at University of Penn. to HPSG. The grammar is one of the grammars with the broadest coverage. Two Japanese grammars, SLUNG and KNP, have also been developed.
(3) Software tools and linguistic resources: Tools for corpus tagging (TMIS) and grammar development (WILL) have been developed to facilitate rapid corpus tagging and grammar development. These tools have been made available to the research community with detailed manuals in English.
(4) Information Extraction from biology texts: In order to show the feasibility of NLP technology, a system of extracting information from scientific texts has been developed by using the systems in (1)-(3). The system was tested by a group of real users. A dialogue system and an intelligent IR systems have also been developed.

11.Key Words

(1)Information retrieval、(2)Information extraction、(3)Knowledge acquisition
(4)Term recognition、(5)Semantic information corpus、(6)Ontology extraction
(7)Intelligent Retrieval System、(8)Discourse analysis、(9)Bioinformatics

12.References

[Reference Articles]
Author Title of Article
Jun-ichi Tsujii Generic NLP Technologies: Language, Knowledge and Information Extraction
Journal Volume Year Pages Concerned
The 38th Annual Meeting of the ACL   2000  

Author Title of Article
Katerina. T. Frantzi, Sophia Ananiadou and Hideki Mima Automatic Recognition of Multi-Word Terms: the C-value/NC-value method
Journal Volume Year Pages Concerned
International Journal on Digital Libraries 3-2 2000 pp.115-130

Author Title of Article
Yusuke Miyao, Takaki Makino, Kentaro Torisawa and Jun'ichi Tsujii The LiLFeS abstract machine and its evaluation with the LinGO grammar
Journal Volume Year Pages Concerned
Journal of Natural Language Engineering, Cambridge University Press 6(1) 2000 pp.47-62

Author Title of Article
Kentaro Torisawa, Kenji Nishida, Yusuke Miyao and Jun'ichi Tsujii An HPSG parser with CFG Filtering
Journal Volume Year Pages Concerned
Journal of Natural Language Engineering, Cambridge University Press 6(1) 2000 pp.63-80

Author Title of Article
Nigel Collier, Chikashi Nobata and Jun-ichi Tsujii Extracting the Names of Genes and Gene Products with a Hidden Markov Model
Journal Volume Year Pages Concerned
18th COLING   2000 pp.201-207

Author Title of Article
Akane Yakushiji, Yuka Tateisi, Yusuke Miyao and Jun-ichi Tsujii Event Extraction From Biomedical Papers Using A Full Parser
Journal Volume Year Pages Concerned
Pacific Symposium on Biocomputing2001   2001 pp.408-419

Author Title of Article
Daisuke Kawahara and Sadao Kurohashi Japanese Case Frame Construction by Coupling the Verb and its Closest Case Component
Journal Volume Year Pages Concerned
Human Language Technology Conference(HLT 2001),   2001  

Author Title of Article
Sadao Kurohashi and Manabu Ori Nonlocal Language Modeling based on Context Co-occurrence Vectors
Journal Volume Year Pages Concerned
The 2000 Joint SIGDAT Conference on Empirical Methods in Natural Language Processing and Very Large Corpora   2000 pp.80-86

Author Title of Article
Sadao Kurohashi and Wataru Higasa Dialogue Helpsystem based on Flexible Matching of User Query with Natural Language Knowledge Base
Journal Volume Year Pages Concerned
1st ACL SIGdial Workshop on Discourse and Dialogue   2000 pp.141-149

Author Title of Article
Sadao Kurohashi, and Yasuyuki Sakai Semantic Analysis of Japanese Noun Phrased : A New Approach to Dictionary-Based Understanding
Journal Volume Year Pages Concerned
37th Annual Meeting of the Association for Computational Linguistics (ACL99)   1999 pp.481-488

Author Title of Article
Sadao Kurohashi and Makoto Nagao Building a Japanese Parsed Corpus while Improving the Parsing System
Journal Volume Year Pages Concerned
The First International Conference on Language Resources & Evaluation   1998 pp.719-724

Author Title of Article
Rila Mandala, Takenobu Tokunaga and Hozumi Tanaka The exploration and analysis of using multiple thesaurus types for query expansion in information retrieval
Journal Volume Year Pages Concerned
Journal of Natural Language Processing 7-2 2000 pp.117-140

Author Title of Article
Rila Mandala, Takenobu Tokunaga and Hozumi Tanaka Query expansion using heterogeneous thesauri
Journal Volume Year Pages Concerned
Information Processing and Management 36-3 2000 pp.361-378

Author Title of Article
Takenobu Tokunaga, Hironori Ogibayashi and Hozumi Tanaka Effectiveness of complex index terms in information retrieval
Journal Volume Year Pages Concerned
The 6th RIAO Conference(RIAO 2000)   2000 pp.1322-1331

Author Title of Article
Makoto Iwayama and Takenobu Tokunaga Probabilistic Passage Categorization and its Application
Journal Volume Year Pages Concerned
Journal of Natural Language Processing 6-3 1999 pp.181-198

Author Title of Article
Makoto Iwayama and Takenobu Tokunaga Associative Document Search using a Probabilistic Document Clustering
Journal Volume Year Pages Concerned
Journal of Natural Language Processing 5-1 1998 pp.101-118

Author Title of Article
Takenobu Tokunaga Computation and Language Volume 5: Information Retrieval and Natural Language Processing
Journal Volume Year Pages Concerned
University of Tokyo Press   1999 234


back