Green Hornet
Green Hornet
Visual analytics for large graphs
(Photo composite by Donald Jorgensen | Pacific Northwest National Laboratory)
About Green Hornet
Graphs and networks are powerful tools used to explore relationships and interactions in complex datasets, enabling analysts to uncover meanings that might otherwise remain hidden. However, visually exploring graphs with more than a few hundred nodes and links can be extremely challenging. To facilitate the exploration of graphs with up to one million nodes and links, Pacific Northwest National Laboratory developed Green Hornet, a user-friendly graph visualization tool. Green Hornet supports large graph exploration and visualization across various domains, including social network analysis, cybersecurity, infrastructure security, and more.
Green Hornet employs a unique multi-scale approach, clustering closely connected nodes into a smaller set of “supernodes” that allow for the interactive exploration of graphs with millions of nodes. Specific individual nodes and edges of interest can be extracted based on their metadata attributes or graph properties, enabling analysts to focus on local relationships of interest while retaining their context within the larger graph.
The tool supports the analysis of various types of graphs and can load data from CSV, GraphML, and GDF files, in addition to its own native format. Graphs can contain diverse properties on both nodes and links, which can be searched as part of the analytics and exploration. Besides graph clustering, Green Hornet also supports other graph analytics capabilities, such as pathfinding and vertex similarity.
Available as both a desktop application and a thin-client web application, the web-based version of Green Hornet now includes integrated Jupyter notebook features. This allows users to interact with Green Hornet via Python in a notebook environment, enabling even more powerful graph analytics capabilities than those available with Green Hornet alone.
Related Publications
- Mackey, P., J. Miller, and L. Faultersack. 2024. “Improving Property Graph Layouts by Leveraging Attribute Similarity for Structurally Equivalent Nodes.” In 2024 IEEE Visualization and Visual Analytics (VIS) (pp. 141-145). IEEE.
- Wong, P. C., D. Haglin., D. Gillen, D. Chavarria, and V. Castellana, C. Joslyn, and S. Zhang. 2015. “A visual analytics paradigm enabling trillion-edge graph exploration.” In 2015 IEEE 5th Symposium on Large Data Analysis and Visualization (LDAV) (pp. 57-64). IEEE.
- Rohrer, R., Paul, C. L., and Nebesh, B. 2014. "Visual analytics for big data." Next Wave, 20, 1-17.
- Wong, P. C., P. Mackey, K. A. Cook, R. M. Rohrer, H. Foote, and M. A. Whiting. 2009. “A multi-level middle-out cross-zooming approach for large graph analytics.” In 2009 IEEE Symposium on Visual Analytics Science and Technology (pp. 147-154). IEEE.
- Wong, P. C., H. Foote, P. Mackey, G. Chin, H. Sofia, and J. Thomas. 2008. “A dynamic multiscale magnifying tool for exploring large sparse graphs.” Information Visualization, 7(2), 105-117.
Contact Us
Questions about Green Hornet? Email us your inquiries.
Patrick Mackey
Nick Cramer
David Gillen
