GRAVITE aims at designing interactive visualization methods and tools
to analyze and mine large datasets. Our emphasis is on the
visualization of graph structures to help users gain insights from
large datasets and large-scale simulations, to understand the data
and/or the underlying model, and ultimately, to identify intrinsic
properties or emergent phenomenon. More than just being able to deal
with large volume and inhomogeneous data, we are required to deal with
constant changes in data, possibly making it ambiguous and uncertain.
In the context of graph visualization, the challenge we face is thus to
design methods and tools:
to deal with large and dynamically changing graphs;
to visually identify salient properties in changing substructures;
identify the multiscale nature of data;
to produce visual cues helping the user to track such changes in situations where dynamic graphs occur.
Our collaborations with experts of other scientific fields as well
as with industry contribute to the overall organization of this
research agenda and serve a twofold objective:
to build theoretical knowledge relevant to information
visualization and visual analytics, and develop a sound methodology for
graph visualization and navigation;
to target transfer opportunities favoring the adoption of our ideas
and technology by other scientific communities and by the industry.
The core strength of our team resides in the development of
combinatorial mathematics and graph algorithmics to serve the aims of
graph visualization. We deploy our mathematical and algorithmic skills
in Information Visualization to develop:
Graph statistics: that capture key properties of the data, including scalable implementations;
Clustering methods: that handle large datasets both visually and computationally;
Graph hierarchies: that transform large graphs into a hierarchy of
smaller, more readable and easier-to-manipulate sub-structures;
Graph drawing algorithms: that lay out large datasets rapidly,
enhancing scalability and addressing domain-specific conventions and
requirements;
Interactions: that exploit graph hierarchies as a central mechanism
for navigating large graphs, while taking domain-specific tasks into
account;
Evaluation methods: that generate artificial datasets (randomly) based on key properties of the target data.