Spring-Based lattice drawing highlighting conceptual similarity

  • Authors:
  • Tim Hannan;Alex Pogel

  • Affiliations:
  • Physical Science Laboratory, New Mexico State University, Las Cruces, NM;Physical Science Laboratory, New Mexico State University, Las Cruces, NM

  • Venue:
  • ICFCA'06 Proceedings of the 4th international conference on Formal Concept Analysis
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents a spring-based lattice drawing method that uses natural spring lengths determined by assigning a dissimilarity value, the size of symmetric difference, to every pair of concept extents. This extends previous work on incorporating support structure in a concept lattice diagram, in which the support weight function was applied to modify any layout. That work was a partial advance toward the goal of viewing high confidence association rules via the lattice diagram in a way that naturally extends the traditional viewing of implications in a diagram, but also caused the appearance of nearly horizontal edges. The spring-based method solves this problem by placing concepts in the ambient space such that the distance between concepts is proportional to the size of the symmetric difference of the extents of the respective concepts. Besides meeting the proportionality criteria, the algorithm yields highly symmetric diagrams in cases where it is expected.