• 研究課題をさがす
  • 研究者をさがす
  • KAKENの使い方
  1. 課題ページに戻る

2020 年度 実績報告書

Archive-based Question Answering

研究課題

研究課題/領域番号 18K19841
研究機関京都大学

研究代表者

Adam Jatowt  京都大学, 情報学研究科, 特定准教授 (00415861)

研究期間 (年度) 2018-06-29 – 2021-03-31
キーワードquestion answering / document archives
研究実績の概要

Our research on question answering resulted in development of datasets and effective methods. The research was published in ECIR2020 conference and in 2021 a journal paper has appeared in Information Retrieval journal in which we extend the dataset and the method to be larger and more generic. Our approach can successfully answer arbitrary user queries for document archives.
We have also edited a book related to computational approaches to language evolution that will appear with Language Science Press and has more than 350 pages.
We have also published three chapters by our team in this book where two of them are surveys: survey of applications of language evolution approaches and survey of language evolution methods as well as one research paper on automatically dividing timeline into non uniform time units for effectively studying the temporal evolution of words.

  • 研究成果

    (3件)

すべて 2021 その他

すべて 国際共同研究 (1件) 雑誌論文 (1件) (うち国際共著 1件、 査読あり 1件) 図書 (1件)

  • [国際共同研究] Karlsruhe Institute of Technology(ドイツ)

    • 国名
      ドイツ
    • 外国機関名
      Karlsruhe Institute of Technology
  • [雑誌論文] Improving question answering for event-focused questions in temporal collections of news articles2021

    • 著者名/発表者名
      Jiexin Wang, Adam Jatowt, Michael Faerber, Masatoshi Yoshikawa:
    • 雑誌名

      Information Retrieval Journal

      巻: 24(1) ページ: 29-54

    • DOI

      10.1007/s10791-020-09387-9

    • 査読あり / 国際共著
  • [図書] Computational approaches to semantic change2021

    • 著者名/発表者名
      Nina Tahmasebi Lars Borin Adam Jatowt Yang Xu Simon Hengchen
    • 総ページ数
      364
    • 出版者
      Language Science Press

URL: 

公開日: 2021-12-27  

サービス概要 検索マニュアル よくある質問 お知らせ 利用規程 科研費による研究の帰属

Powered by NII kakenhi