Social network visualization: Does partial edges affect user comprehension?

Abstract

In a small-world resolution, Euler diagrams are tangentially augmented with graphs and are realized in social network analysis but one of the main issue is the visual clutter. One interesting way of reducing this clutter is to use partial edges when there are visual elements (edge and curve) crossings. This paper empirically evaluates the use of partial edges for a small-scaled Euler diagrams with graphs and tests whether Euler diagrams with partial edges are readable and interpretable. In this study, twenty participants were presented with eighteen manually drawn diagrams using a software which recorded the answers and the time taken to answer each question. The results indicate that there is no significant difference between diagrams having partial and non-partial edges in terms of accuracy and time (performance analysis) but users preferred partial edges to be aesthetically pleasing in terms of reading graphs in Euler diagrams and layout features (preference analysis), which means partial edges can be used as a strategy to avoid clutter. The results obtained from the task performance and preference will be useful in generating real small-world graphs in an optimal way. The paper concludes that partial edges in a graph does not significantly affect comprehension and as a further direction, our existing Euler diagrams with complete graphs (full edges) tool will be extended to partial edges and evaluated.