Dynamic network graphing research project https://rsrch.sysreturn.net
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
James Storey 3fb6bc9e1e Update 'LICENSE' 1 year ago
app bump version for release to 2.30.0 1 year ago
backend combine smaller ego data steps into top level widths 1 year ago
docs revise some task phrasing, update task question docs 1 year ago
lab add tool to query and process metric data 1 year ago
.gitignore revise some docs, start lab project dir 1 year ago
.gitlab-ci.yml add cache directive to ci script 1 year ago
LICENSE Update 'LICENSE' 1 year ago
README.md add link to dev readme in main readme 1 year ago
rsrch.sublime-project combine smaller ego data steps into top level widths 1 year ago


Historical Context in Real-Time Situational Awareness Network Visualizations using Dynamic Ego-Network Graphs


Visual interfaces for information technology networks may use node-link graphs to give an overview of the network's status and topology. To support situational awareness, visualizations need to show both the current state of the network while also keeping the view up to date as time passes. Dynamic graphs can animate over time to allow viewers to monitor network changes, but graph layout can affect the ease of interpreting the graph. The multitude of dimensions needed to be presented provides a major challenge for these graphs. Since two dimensions are used to map vertices and edges, there is no unique dimension for time. Therefore, time is typically shown using animation or the \textit{small-multiples} technique and related methods. Each of these approaches have their own features and trade-offs in granularity and ease of comparison. An important aspect of situational awareness displays is the ability to compare the current state with historical data. However, widely used dynamic graph techniques have trade-offs that are sub-optimal for situational awareness displays and historical data comparison.

We propose using a dynamic ego-centric graphing technique that maps time to both space and time. Doing so will meet the needs of situational awareness analysis more appropriately than other techniques. We will conduct a study to evaluate the efficacy of this technique. Our study will compare a dynamic ego-centric technique against animated and \textit{small-multiple} exo-centric graphs. Participants will be measured on the speed and accuracy of their evaluation across a range of network analysis tasks. The results of this study will help the situational awareness research community by showing how previously unused techniques apply to their problem domain.

Proposal Paper

Application Details

See the application readme