Paper 2: Motif-Based Visual Analysis of Dynamic Network

Eren Cakmak, Johannes Fuchs, Dominik Jäckle, Tobias Schreck, Ulrik Brandes, Daniel Keim

View presentation:2022-10-16T14:52:00ZGMT-0600Change your timezone on the schedule page
2022-10-16T14:52:00Z
Exemplar figure, described by caption below
We propose two complementary scalable pixel visualizations to provide an overview of changing motif structures in large-scale dynamic networks. The network-level census pixel visualization reveals structural changes, trends, states, and outliers. The pipeline displays the steps for generating the network-level census visualization. (1) the network motif significance profile (census) is calculated for each time step, (2) the vectors are presented as a pixel-based visualization, and (3) reordering strategies are used to reveal similar network superfamilies. The reordering strategies are crucial for grouping similar network topologies to emphasize structural changes, trends, states, and outliers.

The live footage of the talk, including the Q&A, can be viewed on the session page, VDS: Opening / Keynote 1 + Paper Session 1.

Abstract