On the Perception of Small Sub-graphs

Auteurs

Miller J., Ghoniem M., Wu H.Y., Purchase H.C.

Référence

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 14465 LNCS, pp. 213-230, 2023

Description

Interpreting a node-link graph is enhanced if similar sub-graphs (or ‘motifs’) are depicted in a similar manner – that is, they have the same visual form. Small motifs within graphs may be perceived to be identical when they are structurally dissimilar, or may be perceived to be dissimilar when they are identical. This issue primarily relates to the Gestalt principle of similarity, but may also include an element of quick, low-level pattern-matching. We believe that if motifs are identical, they should be depicted identically; if they are nearly-identical, they should be depicted nearly-identically. This principle is particularly important in domains where motifs hold meaning and where their identification is important. We identified five small motifs: bi-cliques, cliques, cycles, double-cycles, and stars. For each, we defined visual variations on two dimensions – same or different structure, same or different shape. We conducted a crowd-sourced empirical study to test the perception of similarity of these varied motifs, and found that determining whether motifs are identical or similar is affected by both shape and structure.

Lien

doi:10.1007/978-3-031-49272-3_15