AmbiguityVis: Visualization of Ambiguity in Graph Layouts

摘要: Node-link diagrams provide an intuitive way to explore networks and have inspired a large number of automated graph layout strategies that optimize aesthetic criteria. However, any particular drawing approach cannot fully satisfy all these criteria simultaneously, producing drawings with visual ambiguities can impede the understanding network structure. To bring attention potentially problematic areas present in drawing, this paper presents technique highlights common types ambiguities: ambiguous spatial relationships between nodes edges, overlap community structures, ambiguity edge bundling metanodes. Metrics, including newly proposed metrics for abnormal lengths, structures node/edge aggregation, are quantify drawing. These others then displayed using heatmap-based visualization provides feedback developers approaches, allowing them quickly identify misleading areas. The novel allow user layouts from multiple perspectives order make reasonable choices. effectiveness is demonstrated through case studies expert reviews.