Making provision for Multilingual Information Retrieval

							Making provision for
Multilingual Information Retrieval


                                    Nobuko Miyairi
           LIS670 Information Storage and Retrieval
                                       Spring 2001
   Background
    TUFS (Tokyo University of Foreign Studies)
    – 26 languages
    – students and faculty from other countries

    TUFS Library
    – 500,000 books + 1,000 periodicals
    – over 120 languages
    – 9 librarians



4/30/01                      2
   Challenges
    You don’t know the language
    – “How can I locate web sites about Sufism written
    in Urdu?”

    You don’t know the technology
    – “How can I display Urdu on my computer?”

    You can’t evaluate what you retrieved
    – “the current trend of academic libraries in
    Thailand.”

4/30/01                       3
   Who needs it?
    Information Seeker
    – students, scholars, researchers
    – general audience

    Surrogate Searcher
    – librarians

    Information Provider
    – webmasters, content creators


4/30/01                      4
   Information on WWW
    Growing Scale of WWW
    – more than 1 billion pages (as of 2001)
    – 68% is English


    How about translation?
    – can’t cover the amounts
    – can’t catch up with the speed
    – can’t satisfy diverse information needs


4/30/01                       5
   Solution 1
    Localization
    – OS level
    – mostly bilingual, not multilingual
    Software
    – editor, word processor
    – browser
    – IME (input method editor)

    Gateway Service


4/30/01                        6
   Solution 2
    Tools for Surrogate Searcher
    – cross-lingual search engine


    – automatic translation service



    Standards
    – universal character set (e.g., Unicode, ISO10646)
    – document standard (e.g., XML, MARC21)

4/30/01                       7
   Summary
    Multilingual Computing Technology
    CLIR: Cross-lingual Information Retrieval
    Services on the Web
    Standards

    None of them is cure-all
    – interdisciplinary cooperation
    – and your creativity!


4/30/01                       8