Large graph visualization by hierarchical clustering

This paper proposes a new technique for visualizing large graphs of several ten thousands of vertices and edges. To achieve a graph abstraction, a hierarchical clustered graph is extracted from a general large graph based on the community structures discovered in the graph. An enclosure geometrical partitioning algorithm is then applied to achieving the space optimization. For graph drawing, it uses a combination of spring-embbeder and circular drawing algorithms that archives the goal of optimization of display space and aesthetical niceness. The paper also discusses an interaction mechanism accompanied with the layout solution. The interaction not only allows users to navigate hierarchically through the entire clustered graph, but also provides a way to navigate multiple clusters concurrently. Animation is also implemented to preserve user mental maps during the interaction.