Journal of Information Processing and Management
Online ISSN : 1347-1597
Print ISSN : 0021-7298
ISSN-L : 0021-7298
Search
Advanced search
OR
Browse
Search
Volume 32, Issue 4
Displaying 1-18 of 18 articles from this issue
Show all abstracts
    • |<
    • <
    • 1
    • >
    • >|
Panel Discussion
  • Automated indexing for making of a newspaper article database
    Tatsuo KAMIO
    1989 Volume 32 Issue 4 Pages 283-293
    Published: 1989
    Released on J-STAGE: March 23, 2012
    DOIhttps://doi.org/10.1241/johokanri.32.283
    JOURNAL FREE ACCESS
    Show abstractHide abstract
    Automated indexing has been widely employed in the process of making newspaper article databases. It is essential to speed up the compiling time of the said databases for the large amount of articles come out daily, and save manpower involved in it, with the aid of computers. However, indexed terms which are extracted by the current automated indexing systems have no links with subject analysis, so that they are not considered to be keywords in a strict sense. Thus, the system of Nihon Keizai Shimbun KK enables to justify keywords to certain extent based on the two clues ; 1) at which location the extracted term occurred, and 2) whether or not subject area of the article corresponds to thesaurus class of the extracted term, by using characteristics peculiar to newspaper articles. Also the experiment of assigning keywords which are not occurred in articles was conducted. The fairly good result was obtained.
    View full abstract
    Download PDF (1071K)
Series
    • |<
    • <
    • 1
    • >
    • >|
feedback
 Top