Knowledge Navigation on Visualizing Complementary Documents

  • Authors:
  • Naohiro Matsumura;Yukio Ohsawa;Mitsuru Ishizuka

  • Affiliations:
  • -;-;-

  • Venue:
  • DS '01 Proceedings of the 4th International Conference on Discovery Science
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

It is an up-to-date challenge to get answers for novel questions which nobody has ever considered. Such a question is too rare to be satisfied with a past single document. In this paper, we propose a new framework of knowledge navigation by graphically providing with multiple documents relevant to a user's question. Our implemented system named MACLOD generates several navigational plans, each forming a complementary document-set, not a single document, for navigating a user to understanding a novel question. The obtained plans are mapped into a 2-dimensional interface where documents in each obtained document-set are connected with links in order to support user selecting a plan smoothly. In experiments, the method obtained satisfactory answers to user's unique questions.