Research Article

Constructing an associative concept space for literature-based discovery

  1. C. Christiaan van der Eijk1,
  2. Erik M. van Mulligen1,
  3. Jan A. Kors1,
  4. Barend Mons1,
  5. Jan van den Berg2

Article first published online: 20 JAN 2004

DOI: 10.1002/asi.10392

Issue

Journal of the American Society for Information Science and Technology

Journal of the American Society for Information Science and Technology

Volume 55, Issue 5, pages 436–444, March 2004

Additional Information

How to Cite

van der Eijk, C. C., van Mulligen, E. M., Kors, J. A., Mons, B. and van den Berg, J. (2004), Constructing an associative concept space for literature-based discovery. J. Am. Soc. Inf. Sci., 55: 436–444. doi: 10.1002/asi.10392

Author Information

  1. 1

    Department of Medical Informatics, Erasmus MC—University Medical Center Rotterdam, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands

  2. 2

    Department of Computer Science, Faculty of Economics, Erasmus University Rotterdam, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands

Publication History

  1. Issue published online: 10 FEB 2004
  2. Article first published online: 20 JAN 2004
  3. Manuscript Revised: 26 SEP 2003
  4. Manuscript Accepted: 26 SEP 2003
  5. Manuscript Received: 6 JAN 2003

SEARCH

SEARCH BY CITATION

ARTICLE TOOLS

View Full Article (HTML) Get PDF (241K)

Abstract

Scientific literature is often fragmented, which implies that certain scientific questions can only be answered by combining information from various articles. In this paper, a new algorithm is proposed for finding associations between related concepts present in literature. To this end, concepts are mapped to a multidimensional space by a Hebbian type of learning algorithm using co-occurrence data as input. The resulting concept space allows exploration of the neighborhood of a concept and finding potentially novel relationships between concepts. The obtained information retrieval system is useful for finding literature supporting hypotheses and for discovering previously unknown relationships between concepts. Tests on artificial data show the potential of the proposed methodology. In addition, preliminary tests on a set of Medline abstracts yield promising results.

View Full Article (HTML) Get PDF (241K)