Problems of network simplification by edge bundling

Node occlusion and edge congestion problems, which are caused by the increment of network scale and complexity, had become a hot spot in network visualization research. To solve the visual clutter problem in network, edges close to each other in network were bundled by curving them. The bundling started from node position and group division, laying emphasis on edge bundling technique through edge convergence, edges bundled close to each other in network to reduce visual complexity. A segmental forced directed algorithm (FDA) simplification model and a group based consistent edge bundling network model were proposed and improved. In segmental FDA bundling model, quadratic spline was used to display line in network, control points of spline were produced by iteration to implement edge bundling. To solve the problem of excessive bending of some edges in segmental FDA bundling model, network was divided into different groups by CNM cluster algorithm. Lines in the same group were applied with edge consistent principle on basement of network group structure, and bundling level was calculated by the matching coefficient of edges. After edge bundling, the phenomena of curve's distortion decreased and curves became smoother. Domestic airline network data was chosen as experimental subject, and the network was simplified by two bundling models, and then the simplification result was analyzed. Experimental result shows that positions of nodes in the same group stand close to each other, and airline network shows distinct crossing skeleton. Airlines have north-south and west-east orientation bundle independently, which reveals the whole trends of airline network construction in these two directions. The bundling simplification algorithm introduced mentioned above has a wide applicability, and network visualized by this algorithm has good visual effect and readability.