The State of the Art in Visualizing Multivariate Networks

Abstract

Multivariate networks are made up of nodes and their relationships (links), but also data about those nodes and links as attributes. Most real‐world networks are associated with several attributes, and many analysis tasks depend on analyzing both, relationships and attributes. Visualization of multivariate networks, however, is challenging, especially when both the topology of the network and the attributes need to be considered concurrently. In this state‐of‐the‐art report, we analyze current practices and classify techniques along four axes: layouts, view operations, layout operations, and data operations. We also provide an analysis of tasks specific to multivariate networks and give recommendations for which technique to use in which scenario. Finally, we survey application areas and evaluation methodologies.

Expressive Authoring of Node-Link Diagrams With Graphies

“…However, while expressive tools enable the creation of a wide range of visualizations, they work with tabular data, and have little or no support for network data structures. Those that do, such as iVisDesigner [34] or Charticulator [35], are forced to expose the concepts of nodes and links in an indirect, somewhat cumbersome manner, failing to explicitly represent the multivariate data in the context of the network's topology [31].…”

“…While the above environments make it possible to create a wide range of visualizations, other systems are specifically designed for working with networks and coping with the specific challenges they raise. Indeed, representing the topological structure and the attributes associated with the nodes and links of multivariate graphs is often difficult [31]. For instance, spatial position is typically under the control of the graph layout algorithm, which tries to produce a representation that is legible, based on heuristics such as minimizing link crossings [52].…”

“…As a consequence, node size variations are fairly constrained, eventually limiting the number of channels for the visual mapping of data attributes [37,38]. The problem gets more difficult as graphs grow in size and complexity [31,51] and when they are dynamic [25].…”

“…Beyond the above systems, that are focused on authoring network visualizations, there is also a variety of tools for their visual analysis. We focus here on systems that provide effective support for the visualization of multivariate networks, and refer the reader to Nobre et al's recent survey [31] for a complete picture of the state of the art. One possibility consists in forcing specific layouts by mapping up to two node attributes to spatial position.…”

“…The resulting tools support a wide range of charts, as long as such charts can be derived from tabular data. They offer little or no support for multivariate networks [31], however. Designers who want to create expressive network visualizations must resort to dedicated visualization and analysis software and then customize the output in a general-purpose vector graphics editor.…”