• DocumentCode
    3421987
  • Title

    Automated Japanese essay scoring system:jess

  • Author

    ISHIOKA, Tsunenori ; KAMEDA, Masayuki

  • Author_Institution
    Res. Div., Nat. Center for Univ. Entrance Exam, Tokyo, Japan
  • fYear
    2004
  • fDate
    30 Aug.-3 Sept. 2004
  • Firstpage
    4
  • Lastpage
    8
  • Abstract
    We have developed an automated Japanese essay scoring system named jess. The system evaluates an essay from three features: (1) rhetoric - ease of reading, diversity of vocabulary, percentage of big words (long, difficult words), and percentage of passive sentences; (2) organization - characteristics associated with the orderly presentation of ideas, such as rhetorical features and linguistic cues; (3) contents - vocabulary related to the topic, such as relevant information and precise or specialized vocabulary. The final evaluated score is calculated by deducting from a perfect score assigned by a learning process using editorials and columns from the Mainichi Daily News newspaper. A diagnosis for the essay is also given. Our system does not need any essays graded by human experts.
  • Keywords
    educational administrative data processing; linguistics; natural languages; text analysis; vocabulary; Mainichi Daily News newspaper; automated Japanese essay scoring system; human experts; jess; learning process; linguistic cues; rhetorical features; vocabulary diversity; Automatic testing; Computer errors; Educational technology; Humans; Performance evaluation; Rhetoric; System testing; Technology management; Vocabulary; Writing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Database and Expert Systems Applications, 2004. Proceedings. 15th International Workshop on
  • ISSN
    1529-4188
  • Print_ISBN
    0-7695-2195-9
  • Type

    conf

  • DOI
    10.1109/DEXA.2004.1333440
  • Filename
    1333440