Edinburgh Napier University

  Information Visualisation is extensively used in single graph analysis. However, relatively little work has been done in the field of graph comparison. This work examines and compares the use of two standard graph representations in this area, the Node-Link representation and one based on the graph adjacency matrix. It considers which representation method is superior. In addition it explores whether it is best, for comparison purposes, to combine multiple graphs into single views or to juxtapose single graph representations.
To run this comparison a simple tool was developed and task-based analysis done using that tool to compare multiple versions of a small, locally dense, directed multigraph based on sports data. We are able to demonstrate that it is better to combine views into a single diagram, and that even for small graphs, an analyst is not disadvantaged by the abstract nature of the matrix compared to the intuitive Node-Link diagram.