The Journal of Information Science and Technology Association
(Johono Kagaku to Gijutsu)
Abstract

Kazuaki KISHIDA
Faculty of Letters, Keio University
2-15-45 Mita, Minato-ku, Tokyo 108-8345 JAPAN

Abstract：The purpose of this article is to discuss extensively about effectiveness and functionality of controlled vocabulary, such as thesaurus or subject heading list, in the age of the Internet, which enables us to access a various kinds of full-text information. First, we outline fundamental functions of the controlled vocabulary as a device for enhancing search performance, and also identify its limitation. Next, effectiveness of the controlled vocabulary is discussed in the case of searching academic or technological databases with high specialty and of searching full-text information such as WWW. Finally, we describe briefly possibility of automatic thesaurus construction, ontology, and folksonomy.

Keywords：  controlled vocabulary / thesaurus / subject heading list / retrieval experiment / full text retrieval / ontology

Kyoko UENO
Japan Association for International Chemical Information (JAICI)
Nakai Bldg. 6-25-4 Honkomagome, Bunkyo-ku, Tokyo 113-0021 JAPAN

Abstract：SciFinder, a desk-top research tool for scientists, provides various search capabilities, including structure searching, reaction searching, and keyword searching (Topic Search). This article describes the algorithm behind SciFinder's Topic Search, which helps improve the relevance and comprehensiveness of the search result.

Keywords：  SciFinder / SciFinder Scholar / topic search / algorithm / Boolean operator / proximity operator / synonym

Machie SHIMADA
National Diet Library
1-10-1, Nagatacho, Chiyoda-ku, Tokyo, 100-8924 JAPAN

Abstract：The National Diet Library List of Subject Headings (NDLSH) has had many problems such as shortage of subject headings, lack of "see also" and "see" references, and delay of release. So it has been not easy to use it when searching the NDL-OPAC. The National Diet Library (NDL) has revised NDLSH since 2004 and tried to solve the above problems by adding new subject headings, revising the rules of subject headings and subdivisions, expanding "see also" and "see" references and introducing a thesaurus structure. The NDLSH is now available on the NDL website in PDF file and you can also obtain the text data on request.

Keywords：  Subject headings / NDLSH / subject access / thesaurus / "see also" reference / "see" reference / LCSH / controlled language

Yoshiko TANAHASHI, Nobuko MIYAIRI
Thomson Corporation K.K.
1-1-1, Hitotsubashi, Chiyoda-ku, Tokyo 100-0003 JAPAN

Abstract：The study focuses on the semantic application of cited references in literature displayed as KeyWords Plus and Related Records in the Web of Science database. In a sample search result based on MeSH term "Toll-Like Receptor 10", the frequency of KeyWords Plus and cited references were analyzed. Words and cited items with frequent co-occurrence were combined to retrieve relevant literature for the same topic. Both results showed high precision judged by relevance to the original MeSH term. It was proven that statistical analysis of cited references is effective in selecting meaningful combinations of natural languages used by authors and complements controlled-vocabulary search.

Keywords：  citation / Web of Science / KeyWords Plus / Related Records / controlled-vocabulary / natural language search

Hisamitsu SATO
NTT Advanced Technology Corporation
12-1, Ekimaehoncho, Kawasaki-ku, Kawasaki-shi, Kanagawa Pref. 210-0007 JAPAN

Abstract：Ntt-at Co. has developed "EF/Thesaurus" as a tool to help and improve the method of information retrieval and put it into commercial for more than ten years. To reduce the possibility of information retrieval failure, to select the documents which are retrieved, to help to find the most suitable keyword and to be a useful tool to create thesaurus for individual users are the basic purposes of this thesaurus. In this article, the importance of sticking to synonyms and keeping up to current words are mentioned. The basic rules and the concepts of developing of "EF/Thesaurus" are also explained. Finally, the thesaurus which will be expected in the time of Internet era is mentioned.

Keywords：  information retrieval / keyword / thesaurus / synonym / Internet

Masaaki KASHIMURA
HUMI Project, Keio University
2-15-45 Mita, Minato-ku, Tokyo 108-8345 JAPAN

Abstract：Digital archive is a system for preserving valuable digital information obtained by the digitization of cultural material in a secure and reliable manner and for using it effectively. The HUMI Project has constructed internal digital archives of the digital facsimiles for preservation, and digital archives for web publishing. For the latter, image servers are utilized to show images to users. In the present article, digital archives of rare books and their publication will be elucidated briefly, and some examples will be given of the web publishing of digital archives relating to the HUMI Project.

Keywords：  rare book / digital archive / digital image / digital facsimile / HUMI Project / image server