Research themes in the case-based reasoning in health sciences core literature

  • Authors:

  • Isabelle Bichindaritz

  • Affiliations:

  • State University of New York, Oswego, New York

  • Venue:
  • ICDM'12 Proceedings of the 12th Industrial conference on Advances in Data Mining: applications and theoretical aspects
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Research in case-based reasoning (CBR) in the health sciences started more than 20 years ago and has been steadily expanding during these years. This paper describes the state of the research through an analysis of its mainstream, or core, literature. The methodology followed involves first the definition of a classification and indexing scheme for this research area using a tiered approach to paper categorization based on application domain, purpose of the research, memory organization, reasoning characteristics, and system design. A research theme can be tied to any of the previous classification elements. The paper further analyzes the evolution of the literature, its characteristics in terms of highest impact, or most cited, papers, and draws conclusions from this analysis. Finally, a comparison with the themes automatically learned through clustering co-citations matrices with the Ensemble Non-negative Matrix Factorization (NMF) algorithm in the CBR conference literature is proposed. This comparison helps better understand the main characteristics of the field and propose future directions.