DYNAMIC VISUALIZATION OF GEOGRAPHIC NETWORKS USING SURFACE DEFORMATIONS WITH CONSTRAINTS

This paper proposes a visualization system for geographic network data sets, which aims to convey both low level details of the data and high level contextual information with two different visualization modalities. The first modality, global context visualization represents time-series spatial network data within geographic context through a real-time animation of 3D map deformations. It maintains spatial framework while providing a qualitative feel of the data by only exhibiting dominant and/or interesting features. The second modality provides a set of interactive analytical tools based on conventional node and link displays, which reveal accurate statistical details of the data on demand.

The global context visualization technique employs a modified graph drawing algorithm based on spring-embedders which position network nodes according to the time-series data being fed to it. Our contribution lies in projecting complex network data sets into a real-time virtual environment where the geographic framework is preserved. Applied constraints limit variation of network nodes by favoring inherent geographic distribution of nodes. The graph optimization solution is reached using an implicit integration scheme and allows the system to visualize data in real-time. As the position of network nodes change, the surface is redrawn to fit on the new positions of the nodes. The geographic map projected on the surface deforms and enables viewers to read the data variation as a map deformation. This representation gives a strong qualitative impression and enables viewers to summarize the nature of the data.

The first phase of the global context visualization employs spring-embedders for drawing a graph in which each geographic location corresponds to a single node and non-spatial data components correspond to relations between these nodes. The output graph visualization reflects input data by positioning related nodes closer. In this sense, our technique can be compared to force-directed placement methods. However, the proposed technique does not follow force-directed placement in any precise sense, but instead exploits its key features. The single most important distinction lies in the geographic constraints applied on the system. These constraints enable the spring-embedder system to reach a configuration that will lead to a deformed map where geographic layout is preserved to some extent for assuring intuitive recognition.

In the second phase, the surface covering the nodes is adjusted to fit on the modified positions. As a result, the geographic map projected on the surface deforms and highlights variations in the data. This approach exploits a priori knowledge of the viewer about the physically accurate version of the map. In other words, map deformation facilitates comprehension of non-spatial variables with respect to the geographic framework. Once users have a sense of the overview of the data, they can dig into the details by using analytical tools provided. Interactively responding to users, these tools expand informative quality of the visualization through direct manipulation of visualization parameters. They can explore network data as height bar animations over 3D map or as arcs showing connections. They are able to filter data through selecting nodes or selecting the range of the displayed data.

We examined the proposed method using two different data sets. The first data set is the domestic air flights of US among 231 airports between 1991 and 2004. The second data set is the diplomatic exchange data among 128 nations through years 1815 to 1966.