NTCIR Workshop 2
Proceedings of the Second NTCIR Workshop on Research in Chinese & Japanese Text Retrieval and Text Summarization
May 2000- March 2001


National Institute of Informatics, Tokyo, Japan
Copyright (C) 2001 National Institute of Informatics
ISBN: 4-924600-96-2
Organized by:NII  (National Institute of Informatics)

In cooperation with:
National Taiwan University
IPSJ (Information Processing Society of Japan)
SIG-FI (Special Interest Group on Fundamental Infology), IPSJ

Supported by:
JSPS (Japan Society for the Promotion of Science) "Research for the Future Program: Studies on Ubiquitous Information Systems for Utilization of Highly Distributed Information Resources" (Principal Investigator: Jun Adachi, Professor, NII)


Proceedings for other NTCIR workshops can be found at the NTCIR proceedings page.


Table of contents

Preface from the Organizing Chair [PDF]
Jun Adachi (National Institute of Informatics)

Preface from the Program Chair [PDF]
Noriko Kando (National Institute of Informatics)


1. Keynote Speech

"On Laboratory Testing of Text Retrieval Systems" [PDF]
Stephan Robertson (Microsoft Research Cambridge,U.K. & City University,London, U.K.)


2. Invited Speech

"The Importance of Focused Evaluations: a Case Study of TREC and DUC" [PDF]
Donna Harman (Retrieval Group, Imformation Access Division, Information Technology Laboratory, National Institute of Standards and Technology)
"Evaluation as Enabling Tool for Research and Development" [PDF]
Daniel Marcu(Information Sciences Institute, University of Southern California)


3. Panel

3.1 Panel I:
"Evaluation of Multilingual Information Access; Evaluation Methods & Metrics, Resources, & International Collaboration", Coodinator: Douglas W. Oard (University of Maryland)
Sung Hyon Myaeng (Department of Computer Science, Chungnam National University)
Martin Braschler (Eurospider IT AG)
Fredric C. Gey (UC Data Archive &Technical Assistance, University of California at Berkeley)
Douglas W. Oard (University of Maryland)
Du Lin (Institute of Software, Chinese Academy of Sciences)
Atsushi Fujii (University of Library and Information Science)

The CLEF Campaign [PDF]
Martin Braschler* and Carol Peters** (*Eurospider IT AG, **IEI-CNR)

3.2 Panel II:
"Way Ahead of IR &Summuraization Research", Coordinator: Noriko Kando (National Institute of Informatics)
Daniel Marcu (Information Sciences Institute, University of Southern California)
Donna Harman (National Institute of Standards and Technology)
Mun-Kew Leong (BIGontheNet Pte. Ltd.)
Takahiro Fukusima (Otemon Gakuin University)
Jun'ichi Fukumoto (Department of Computer Science, RitsumeikanUniversity)
Summarization Evaluation: An Overview [PDF]
Inderjeet Mani (The MITRE Corporation)
Patent Data For IR Research and Evaluation [PDF]
Mun-Kew Leong (BIGontheNet Pte. Ltd.)


4. Overview

4.1 NTCIR Workshop 2
Overview of the Second NTCIR Workshop [PDF]
Noriko Kando (National Institute of Informatics)
4.2 Text Summarization Task
Text Summarization Challenge: Text Summarization Evaluation at NTCIR Workshop2 [PDF]
Takahiro Fukusima* and Manabu Okumura** (*Otemon Gakuin University, **Tokyo Institute of Technology)
4.3 Chinese Information Retieval Task
The Chinese Text Retrieval Tasks of NTCIR Workshop 2 [PDF]
Kuang-hua Chen and Hsin-Hsi Chen (National Taiwan University)
4.4 Japanese & English Information Retrieval Task
Overview of Japanese and English Imformation Retrieval Tasks (JEIR) at the Second NTCIR Workshop [PDF]
Noriko Kando, Kazuko Kuriyama and Masaharu Yoshioka (National Institute of Informatics)


5. Research Papers

5.1 Chinese Information Retrieval Task
NTCIR-2 ECIR Experiments at Maryland: Comparing Pirkola's Structured Queries and Balanced Translation [PDF]
Douglas W. Oard and Jianqiang Wang (University of Maryland)
Trans-EZ at NTCIR-2 : Synset Co-occurrence Method for English-Chinese Cross-Lingual Information Retrieval [PDF]
Guo-Wei Bian and Chi-Ching Lin (Trans-EZ Information Technology Inc.)
NTCIR-2 Chinese, Cross Language Retrieval Experiments Using PIRCS [PDF]
K. L. Kwok(Queens College, City University of New York)
CRL at NTCIR2 [PDF]
Masaki Murata, Masao Utiyama, Qing Ma, Hiromi Ozaku and Hitoshi Isahara (Communications Research Laboratory)
Hybrid Term Indexing: an Evaluation [PDF]
Robert W.P.Luk*, K.F.Wong** and K.L.Kwok*** (*Hong Kong Polytechnic University, **Chinese University of Hong Kong,***Queen's College, City University of New York)
Berkeley at NTCIR-2: Chinese, Japanese, and English IR experiments [PDF]
Aitao Chen*, Fredric C. Gey** and Hailing Jiang* (*School of Information Management and Systems, University of California at Berkeley, **UC Data Archive & Technical Assistance, University of California at Berkeley)
ISCAS: Text Retrieval in NTCIR Workshop II [PDF]
Zhang Yibo, Sun Le, Du Lin, Jin Youbing and Sun Yufang (Institute of Software, Chinese Academy of Sciences)
Nathu IR System at NTCIR-II [PDF]
Jason Chang, David Yu, Ching Ting Shen, Afra Cheng, Garfield Shen, Giordano Shen and David Wong (National Tsing Hua University)
Rerank Method Based on Individual Thesaurus [PDF]
Qu Youli, Xu Guowei and Wang Jun(Fujitsu R & D Center Co.,Ltd.)
5.2 Japanese & English Information Retrieval Task
Flexible Pseudo-Relevance Feedback for NTCIR-2 [PDF]
Tetsuya Sakai*, Stephen E. Robertson**/*** and Stephen Walker** (*University of Cambridge / Toshiba Corporate R &D Center, **Microsoft Research Ltd, ***City University, London)
Cross-Lingual Information Retrieval based on LSI with Multiple Word Spaces [PDF]
Tatsunori Mori, Tomoharu Kokubu and Takashi Tanaka (Division of Electrical and Computer Engineering, Yokohama National University)
Notes on the Limits of CLIR Effectiveness: NTCIR-2 Evaluation Experiments at Justsystem [PDF]
Sumio Fujita (JUSTSYSTEM Corporation)
Regression Model and Query Expansion for NTCIR-2 Ad Hoc Retrieval Task [PDF]
Kazuaki Kishida (Faculty of Cultural Information Resources, Surugadai University)
The Effect of Document Clustering in Interactive Relecance Feedback [PDF]
Makoto Iwayama*, Yoshiki Niwa*, Shingo Nishioka*, Akihiko Takano**, Toru Hisamitsu*, Osamu Imaichi*, Hirofumi Sakurai*, and Masakazu Fujio*(*Central Research Lab., Hitachi, Ltd., **National Institute of Informatics)
R2D2 at NTCIR 2 Ad-hoc Task: Relevance-based Superimposition Model for IR [PDF]
Teruhito Kanazawa, Atsuhiro Takasu and Jun Adachi (University of Tokyo)
NTCIR Experiments Using the OASIS System [PDF]
Vitaliy Kluev*, Mikhail Bessonov** and Vladimir Dobrynin*** (*The University of Aizu, **Universitaet Tuebingen, ***Saint Petersburg State University)
Empirical Term Weighting [PDF]
Kyoji Umemura, Yoshiyuki Takeda, Michiko Tanaka, Li Feng and Eiko Yamamoto (Toyohashi University of Technology)
On a Retrieval Support System by Suggesting Terms to a User [PDF]
Hiroyuki Sakai, Kiyonori Ohtake and Shigeru Masuyama (Toyohashi University of Technology)
RICOH at NTCIR-2 [PDF]
Yasushi Ogawa and Hiroko Mano (Software Research Center, RICOH Co., Ltd.)
NTCIR-2 Experiments Using Long Gram Based Indices [PDF]
Takashi Sato, Nao Hatta, Koji Hiraiwa, Kihei Kobata, Akihiro Furusho and Koto Han (Osaka Kyoiku University)
The Intelligent Method of Information Retrieval Based on Self Organized Knowledge Resources [PDF]
Takayuki Morimoto*, Takahiro Kondo**,Katsuhiko Sugita**, Daisuke Ishikawa**, Masaya Ikemura** and Yuzuru Fujiwara*** (*Faculty of Science, Kanagwa University, **Graduate School of Science, Kanagawa University, ***National Center For Industrial Property Information)
Japanese and English Cross-lingual Information Retrieval at DLUT [PDF]
Seigo Tanimura*, Masashi Suzuki**, HiroshiNakagawa* and Tatsunori Mori** (*Information Technology Center, The University of Tokyo, **Faculty of Engineering, Yokohama National University)
Evaluationg Multi-lingual Information Retrieval and Clustering at ULIS [PDF]
Atsushi Fujii*/** and Tetsuya Ishikawa* (*University of Library and Information Science, **CREST, Japan Science and Technology Corporation)
Structured Index System at NTCIR Workshop 2: Information Retrieval Methods Using Ordered Co-occurrence of Words and their Dependency Relationships [PDF]
Atsushi Matsumura, Atsuhiro Takasu and Jun Adachi (National Institute of Informatics)
Experiments in the Retrieval of Unsegmented Japanese Text at the NTCIR-2 Workshop [PDF]
Paul McNamee (Johns Hopkins University Applied Physics Laboratory)
Document Retrieval in Consideration of the Amount of Term Frequencies [PDF]
Hiroshi Umemoto, Tadanobu Miyauchi and Yoshihiro Ueda (Fuji Xerox Co., Ltd.)
Information Retrieval using Relevance Feedback [PDF]
Shuntaro Isogai, Shigeki Ohira and Katsuhiko Shirai (School of Science and Engineering, Waseda University)
NTCIR-2 Experiments at Matsushita: Monolingual and Cross-Lingual IR Tasks [PDF]
Mitsuhiro Sato and Naohiko Noguchi (Multimedia Systems Research Laboratory, Matsushita Electric Industrial Co., Ltd.)
Approximate Dimension Reduction at NTCIR [PDF]
Fan Jiang and Michael L. Littman (Department of Computer Science, Duke University)
Analysis of the Usage of Japanese Segmented Texts in NTCIR Workshop 2 [PDF]
Masaharu Yoshioka, Kazuko Kuriyama and Noriko Kando (National Institute of Informatics)
The Effect of Cross-Lingual Pooling on Evaluation [PDF]
Kazuko Kuriyama, Msaharu Yoshioka and Noriko Kando (National Institute of Informatics)
5.3 Text Summarization Task
Term Weighting Method based on Information Gain Ratio for Summarizing Documents Retrieved by IR Systems [PDF]
Tatsunori Mori, Miwa Kikuchi and Kazufumi Yoshida (Division of Electrical and Computer Engineering, Yokohama National University)
Sentence Extraction System Assembling Multiple Evidence [PDF]
Chikashi Nobata*, Satoshi Sekine**, Masaki Murata*, Kiyotaka Uchimoto*, Masao Utiyama* and Hitoshi Isahara* (*Communications Research Laboratory, **New York University)
Hybrid Text Summarization Method based on the TF Method and the Lead Method [PDF]
Kai Ishikawa, Shinichi Ando and Akitoshi Okumura (C & C Media Research, NEC Corporation)
Yet Another Summarization System with Two Modules using Empirical Knowledge [PDF]
Kiyonori Otake, Daigo Okamoto, Mitsuru Kodama and Shigeru Masuyama (Toyohasi University of Technology)
How Small a Distinction among Summaries can the Evaluation Method Identify? [PDF]
Yoshio Nakao (Fujitsu Laboratories Ltd.)
Text Summarization based on Hanning Window and Dependency Strucuture Analysis [PDF]
Tsutomu Hirao*, Mamiko Hatayama*, Satoshi Yamada** and Kazuhiro Takeuchi** (*NTT Communication Science Laboratories, **Nara Institute of Science and Technology)
Modified Key-Sentence Extraction by RICOH at NTCIR-2 TSC [PDF]
Masayuki Kameda (Software Research Center,RICOH Co., Ltd.)
Phrase-representation Summarization Method and Its Evaluation [PDF]
Mamiko Oka and Yoshihiro Ueda (Fuji Xerox Co., Ltd.)
A System for Text Summarization Based on Word Importance Measures [PDF]
Hiroshi Ishii, Rihua Lin and Teiji Furugori (Department of computer Science, The University of Electro-Communications)


6. Proposal for the next NTCIR Workshop

An Overview of Question and Answering Challenge (QAC) of the Next NTCIR Workshop [PDF]
Jun'ichi Fukumoto* and Tsuneaki Kato** (*Department of Computer Science, Ritsumeikan University,**Department of Language and Information Science, Univerisity of Tokyo)



NTCIR Workshop 2 Meeting
Proceedings of the Second NTCIR Workshop Meeting on evaluation of Chainese & Japanese Text Retrieval and Text Summarization
March 7-9, 2001


National Institute of Informatics, Tokyo, Japan
Copyright (C) 2001 National Institute of Informatics
ISBN: 4-924600-89-X
Organized by: NII (National Institute of Informatics)

In cooperation with:
National Taiwan University
IPSJ (Information Processing Society of Japan)
SIG-FI (Special Interest Group on Fundamental Infology), IPSJ

Supported by:
JSPS (Japan Society for the Promotion of Science) "Research for the Future Program: Studies on Ubiquitous Information Systems for Utilization of Highly Distributed Information Resources" (Principal Investigator: Jun Adachi, Professor, NII)



6. Evaluation Results
6.1 Chinese Information Retrieval Task: Evaluation Results of Each Run
Chinese Monolingual IR (CHIR)
English-Chinese IR(ECIR)
6.2 Japanese & English Information Retrieval Task
6.2.1 List of Submitted Runs
6.2.2 Evaluation Results of Each RunS
Notes on Evaluation for Japanese &English IR Tasks
---- Monolingual IR ----
Retrieval of Japanese documents using Japanese search topics. J-J [tar ball]
Retrieval of English documents using English topics. E-E [tar ball]
---- Cross-Lingual IR ----
Retrieval of Japanese documents using English topics. J-E [tar ball]
Retrieval of English documents using Japanese topics. E-J [tar ball]
Retrieval from a collection containing a mixture of Japanese and English documents using either Japanese topics. J-JE [tar ball]
Retrieval from a colletion containing a mixture of Japanese and English documents using either English topics. E-JE [tar ball]
6.2.3 System Description Forms
Japanese Topics
Japanese Topics(Segmented)
English Topics
6.3 Text Summarization Task: Evaluation Results of Text Summarization Challenge

top