Embedding, Clustering and Coloring for Dynamic Maps

Abstract

We describe a practical approach for visualizing multiple relationships defined on the same dataset using a geographic map metaphor, where clusters of nodes form countries and neighboring countries correspond to nearby clusters. Our aim is to provide a visualization that allows us to compare two or more such maps (showing an evolving dynamic process, or obtained using different relationships). In the case where we are considering multiple relationships, e.g., different similarity metrics, we also provide an interactive tool to visually explore the effect of combining two or more such relationships. Our method ensures good readability and mental map preservation, based on dynamic node placement with node stability, dynamic clustering with cluster stability, and dynamic coloring with color stability.

Egocentric storylines for visual analysis of large dynamic graphs

“…The clustering and relevance calculation can be computationally expensive, and it would be beneficial for them to be replaced with alternatives that could handle incremental changes. The heuristic approach of Hu et al might be a good place to start [17]. Then the approach might even be efficient enough for real-time analyses.…”

“…A major issue for both methods is to ensure the stability of the layout [19], [13], [5], [17]. A stable layout helps preserve the user's mental map as there is less movement between time-steps, but sacrifices quality in terms of readability for later time-steps as their layout depends on previous time-steps.…”

“…Hu et al [17] proposed a method based on a geographical metaphor to visualize a summary of clustered dynamic graphs. They also use a clustering approach similar to ours, though it sacrifices some local quality for calculation efficiency and additional temporal stability.…”

A Taxonomy and Survey of Dynamic Graph Visualization

“…[MKH12] and Hu et al . [HKV12] generate more space‐filling drawings with directly bordering ‘countries’. Abello et al .…”

Visualizing the Evolution of Communities in Dynamic Graphs

“…Dynamic graphs are often visualized using animation, timeline‐based static visualizations [BBDW14] or hybrid approaches such as DiffAni, which combines the first two with a difference representation [RM13] Timeline‐based approaches, so far, have only looked at the transition of graphs and considered the community structure as stable [BVB*11, GBD09]. In contrast, for animated diagrams, the evolution of communities has already been visualized: in addition to colour‐coding vertices, the drawing space can be partitioned into (not necessarily connected) regions [HKV12, MKH12], convex (potentially overlapping) shapes [FT04, KG06] or nested boxes displaying a cluster hierarchy [MELS95, RPD09]. However, animation can lead to high cognitive load [APP11]; it is difficult to follow multiple community transitions happening at the same time and to track communities or individuals across longer periods.…”