TimeMatrix: Visual Representation for Temporal Pattern Detection in Dynamic Networks, VAST 2018 Mini-Challenge 3

This work proposes a visual analytic technique for visualizing temporal pattern detection in networks. The force-directed layout is a popular way to highlight structures of a network however it does not allow to present the dynamic features (how communities change over time). We target this problem by forcing nodes to display vertically and using the horizontal space for representing the timeline of entities. To show the connection between nodes, the edges are drawn vertically to avoid edge-crossings. This work also presents the result of applying the solution to the VAST 2018 - Mini Challenge 3 dataset, which led to the Honorable Mention: Representation of Small-Scale Temporal Patterns for the challenge.