Visualising small world graphs using agglomerative clustering around nodes of interest

Abstract

The difficulty of visualising large graphs lies not just in processing power and display size but in the inherent visual complexity of a large data-set, as the noise and clutter from large numbers of nodes and an order of magnitude more of edges negatively impacts the comprehensibility of any visualisation. Small world graphs are a classification of graph that occurs frequently in models of real world networks such as computer systems and social networks. The overall objective of our research is to allow users to get a better comprehension of the relationships between data entities in the visualisation of real world systems.