Measuring and improving the readability of network visualizations

Опубликовано 28 июля 2016, 1:25

Network data structures have been used extensively in recent years for modeling entities and their ties for many diverse disciplines. Analyzing networks involves understanding the complex relationships between entities as well as any attributes, statistics, or groupings associated with them. The omnipresent node-link visualization excels at showing network topology and features simultaneously, but many node-link visualizations are not easily readable or difficult to extract meaning from because of inherent network complexity or size. Moreover, for every network there are many potential unintelligible or even misleading drawings. In this talk we discuss three strategies to help users create more effective node-link visualizations in NodeXL, as well as a novel alternate network visualization technique called GraphTrail. We first detail readability metrics to quantify the effectiveness of node-link visualizations, localize areas needing improvement, and feed into assistive layout tools. Next, we introduce a technique called motif simplification that leverages the repeating patterns or motifs in a network to reduce visual complexity and increase readability. We then discuss meta-layout algorithms that take attribute- or topology-based groupings into account, so as to more clearly show the ties within groups and the aggregate relationships between groups. Moving beyond node-link visualizations, we turn to GraphTrail, a technique for analyzing networks through exploration of node and edge aggregates. GraphTrail captures users' interactions and integrates this history directly in the exploration workspace, which improves exploration recall and sharing of analyses with others.