Adaptive Visualization of Social Networks (ADVIS)

This proposal is to develop a new system for Adaptive Visualization of Social Networks (ADVIS). Numerous social network visualization systems have been produced in recent years. But even in the few top systems cognitive science and design principles have been applied in an ad hoc and non-systemic manner. Accordingly, each visualization system is a one-of, which may be fine-tuned to a particular user goal and data set, but which can not adapt to rapidly changing user goals, data sets and visualization techniques. In our new approach, visualizations will be automatically generated based on user-defined goals, data set characteristics, and first principles in cognitive science. This will be a systematic and rigorous process, in which various visualization techniques are evaluated based on their utility for the current user goals and data set, as well as the compatibility of their underlying cognitive premises. The key innovation in our system is the explicit inclusion of all of these factors in an automated design process through the use of taxonomies for cognitive principles, visualization techniques, and user-goals. ADVIS will permit user-goals and visualization techniques to be specified and added to the system, so that optimal visualizations can be rapidly produced.