Visualizing Network Attribute Evolution

Jump to: Project Description | Videos | Publications | Participants | Related Projects

Project Description

The goal of this project is to provide overviews of network attribute evolution while allowing analysts to dive deep into network attributes in any given time period. Rather than a conventional node-link diagram or graph, NetVisia (previously called "NetFlow") accomplishes its goals by showing a heat map of node attributes. Attributes include both standard network metrics as well as content based metrics. Standard values include such metrics as degree and betweenness centrality. Content metrics include the ability to search for frequency of terms, values or custom attributes for nodes and edges. Examples include certain keywords in a co-authorship social network or OSPF neighbors in a computer network. NetVisia is designed to work with any type of network: computer network, social network, etc.

There are two visual areas of NetVisia, the Overview Workspace Window and the two Temporal Windows. The first includes the overview heat map of the network. The layout of the heat map is as such:

  • time is the horizontal axis,
  • nodes are arranged vertically, and
  • node attribute values are coded white (low) to red (high). Black indicates a node is not present in the network at that time point.

NetVisia has the ability to cluster and align nodes. In the visualization below, nodes are aligned based on the first time they enter the network and like nodes are clustered.

The temporal windows allow the user to perform a deep dive on a time period highlighted in the Workspace Window. In this space we have a listing of nodes and node-to-node adjacency matrices. The listing contains all metrics for each node in the selected time period above in the Workspace Window. The adjacency matrices show a similar heat map as the Workspace Window, however the metrics listed are edge-centric unlike the Workspace Window with is node-centric. The adjacency matrices include the following metrics:

  • degree between nodes
  • cosine similarity
  • Pearson Coefficients
  • User-define content metrics

Video Demonstration

NetVisia Overview


NetVisia is still under development, so we have not made NetVisia publicly available yet. However, if you have any dataset that you think NetVisia can help you analyze them, we will be happy to discuss the possibilities and work with you. Please contact us for more information.


Robert Gove, Nick Gramsky, Rose Kirby, Emre Sefer, Awalin Sopan, Cody Dunne, Ben Shneiderman, & Meirav Taieb-Maimon "NetVisia: Heat Map and Matrix Visualization of Dynamic Social Network Statistics and Content," submitted for conference review.

Robert Gove, Nick Gramsky, Rose Kirby, & Emre Sefer, "Analyzing Evolution of Network Attributes and Content with NetFlow: A Network Evolution Visualization Tool," project report.

Other Presentations

"NetFlow: Visualizing Network Attribute Evolution," HCIL 28th Annual Symposium, College Park, MD, 2011.


Web Accessibility