[To Research Bulletin of the NACSIS page]  [To NII publication page]  [To NIIJournal page](Japanese)

Research Bulletin of the NACSIS, No.2 (1989.3)

Preface (Japanese)

1. Takashi HAMADA (National Center for Sience Information System), T. YAMAZAKI (University of Tokyo)
A Support System for Software Development based on Program Database (Japanese)

A support system for software development using a database of function modules is described. There are three levels of an expression for each function module, that is, a samantic expression to be used for retrieval, a function specification that is independent of specific programming language, and a source prpgram. Program modules can be retrieved to be reused for composing another function, and the general specification is translated into a source program.

2. Shoichiro ASANO (National Center for Sience Information System), Yasuaki SASAKI, Choi Myun Heung (University of Tokyo)
A study on high speed packet switching by self routing control (Japanese)

This paper proposes a high speed and high capacity packet switching system by self routing control, which accommodates various communication services, such as data, voice and video. This system is composed of Benes network based on newly proposed routing scheme including transmission strategy. The paper also shows that the proposed system features low packet loss due to collision on the switch, a little delay time as well as adequate amount of hardware. The proposed system will realize a high performance packet switching system which has a through-put of 100 G bits per second.

3. Eisuke NAITO (National Center for Sience Information System)
Problems of UNIMARC-authority (Japanese)

4. Masamitsu NEGISHI (National Center for Sience Information System)
Preparation and Telecommunication of Machine-Readable Documents in Japanese Academic Community (Japanese)

The present state regarding preparation of machine readable documents and transmission of those by telecommunication at Janpanese academic community is summarized in contrast with the general situation formally depicted in the governmental white papers. A reproduction model for scientifc information is proposed to lead the discussion which follows. As "Science Information System" is the basic concept governmentally approved to develop an efficient information distribution system for the university people, computer networks have now been constructed within and among universities. Preparation of scientific documents in machine readable forms and creation of databases are the next topics to be discussed, where the current practice in utilizing Japanese word processors, facsimile, desktop publishing, full text detabases, etc. is touched upon. Standadization aspect for document files is discussed along with the full text database project at the National Center for Science Information System (NACSIS).
The paper was prepared for the seminar, "Telecommunication in business and science", Nov. 30-Dec. 2, 1988, at Japanese-German Center Berlin (Japanisch-Deutches Zentrum Berlin), West Berlin, and was presented as a report from Japanese delegation.

5. Masashi MAKIMURA (Nagoya University Library), Masamitsu NEGISHI (National Center for Sience Information System)
Development of NACSIS-IR Information Retrieval System for ISTP&B (Index to Scientific and Technical Proceedings and Books) Database (Japanese)

NACSIS: National Center for Science Information System has begun an information retrieval service "NACSIS-IR" for university researchers and librarians in 1987. The paper describes a method devised in incorporation of ISTP&B database into NACSIS-IR, which contains information of monograph level and individual paper level.
Conventional information retrieval systems do not provide functions to serch and display records with hierarchical structure as seen in ISTP&B. A design of the online database developed here, providing users with the facility to easily locate and output data on monographs and papers, will be applicable to conventional information retrieval software packages. The paper also touches upon the problems in introducing bibliographic databases with hierarchical record organization into DBMS software and information retrieval packages.

6. Teruo KOYAMA (National Center for Sience Information System)
A study for structural analysis of Japanese medical terms (Japanese)

In variety of research fields, scientific terms are important for information retrieval systems. To attain proper retrieval, the relationship among terms should be administrated systematically. Because most of scientific terms represent composit concepts, it is prospective that the administration can be made efficient, if terms can be decomposed to the elementary concepts. Japanese terms can be regarded relatively easy for these kind of decomposition. In this paper, the author introduces a preliminary study about structual analysis of Japanese medical terms. A morphological analysis method, the internal structure of a computerised dictionary for quick retrieval of elementary terms, consideration about some rules constructing composit terms from elementary ones are discussed.

7. Keizo OYAMA (National Center for Sience Information System)
Database construction of the scientific terms (Japanese)

As processes of database construction of the scientific terms, the author closely examined almost all the existing books titled "Japanese Scientific Terms" in various fields, designed the data formats to store informations in the database and the procedures to convert both printed and CTS-format data sources into the database input format, and actually carried out the procedures. As the result, various features of the informations accompanied with each term and the method to represent term are made clear. The author also describes the proof-reading method using a UNIX-based engineering workstation and its effects.

8. Kyo KAGEURA (National Center for Sience Information System)
Recent Probrems in String Indexing Systems : An Analysis of PRECIS (Japanese)

This paper summerizes the state of the art and some recent problems in string idexing systems and makes clear the cause of problems by analysing PRECIS index data. The possible contribution of string indexing systems to the problem of subject access is then considered.


Patricia M. BATTIN (Columbia University)
Libraries, Technology and Scholarship : The Library as an Information Center (English)