Visual Analysis of Network Centralities (Hong/Xu)

Centrality is the most significant concept for such networks; it identifies the most important node in the network. For example, in biological networks such as protein-protein interaction networks, one can identify the most essential protein. In social networks such as the Hollywood movie actor collaboration network, one can identify the most prominent actor.

A particularly important task is the comparison of different centrality measures within one network. We design three new methods for the exploration and comparison of centrality measures within a network:

3D parallel coordinates: nodes are placed on vertical lines
Orbit-based comparison: nodes are placed on concentric circles
Hierarchy-based comparison: nodes are placed on horizontal lines

For each centrality measure, the graph is copied and drawn on a separate 2D plane with node position dependent on centrality.  These planes are then stacked into the third dimension so that the different centrality measures may be easily compared. 


[Hierarchy-Based Comparison: Biological Network]

These methods are implemented as a part of GEOMI, a visual analytic tool being developed by the VALACON project.